Google Patents Staple of '70s Mainframe Computing

theodp writes "'The lack of interest, the disdain for history is what makes computing not-quite-a-field,' Alan Kay once lamented. And so it should come as no surprise that the USPTO granted Google a patent Tuesday for the Automatic Deletion of Temporary Files, perhaps unaware that the search giant's claimed invention is essentially a somewhat kludgy variation on file expiration processing, a staple of circa-1970 IBM mainframe computing and subsequent disk management software. From Google's 2013 patent: 'A path name for a file system directory can be "C:temp\12-1-1999\" to indicate that files contained within the file system directory will expire on Dec. 1, 1999.' From Judith Rattenbury's 1971 Introduction to the IBM 360 computer and OS/JCL: 'EXPDT=70365 With this expiration date specified, the data set will not be scratched or overwritten without special operator action until the 365th day of 1970.' Hey, things are new if you've never seen them before!"

Comments

Really!

[jhb146]

This is supposed to be new...

Re:Really!

[Anonymous Coward]

Let alone patentable? I remember when the idiots at the USPTO gave a patent on the x-oring of a byte of screen memory to flash a cursor.

Too bad we couldn't generate electricity from stupidity. We seem to have plenty to go around these days.

Re:Really!

[rtfa-troll]

Yes; maybe; and the whole summary is stupid. From claim one of the patent; the very first paragraph:

having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers

So; where the mainframes of the 70s had single consolidated disks stores this is talking about doing this on a distributed filesystem. The area of application is indeed new completely opposite to the claim of the summary.

Patents are not supposed to control what you do; instead they control how you do it. Since the way that Google is claiming to do this is by going around comparing the timestamps on a bunch of different distributed chunks of a file, this is something that no mainframe of the 70s is likely to have had to to so it may even be a new way to automatically delete temporary files. I wish people would begin to understand this and commenters would point it out every time. I wonder if this isn't a bunch of patent lawyers trying to make us look silly.

Having said that; If you had a distributed file which kept a timestamp on each of several separate chunks, how would you go about deciding when to automatically delete it? My guess is that the solution you would come up with quickly is basically the one in the patent. You certainly wouldn't have great difficulty in deciding how you do it; suddenly think "maybe there's a patent that might tell me how to do this"; go to the patent office and read the patent then come back inspired and manage to solve your problem only because Google was so nice as to publish their solution. Patents are supposed to record valuable secrets that companies might otherwise keep to themselves in a way that helps humanity. This one is failing at that.

What this comes down to is that the whole idea of patents on things as abstract as software is stupid and is an illegal interference in free speech a right everyone should have under the universal declaration of human rights. The patent officers of the USPTO and the congressmen who put them there should be arrested.

Re:Really!

[sjames]

The 'plurality' of chunks is irrelevant to the matter of expiring old data. What matters is that you have some sort of metadata telling the system when a blob of data that might resemble a file is to be deleted.

The USPTO, meanwhile, keeps making me feel like blowing a plurality of chunks into a water filled receptacle where upon manipulation of a lever or chain fresh water will enter the receptacle transferring momentum to the relevant chunks causing them to exit through a drain.

Re:Really!

[Alioth]

It's not prior art, it's obviousness. In terms of file storage, I consider myself "ordinarily skilled in the art", yet 5 years ago I put in such a system to expire files at work on a distributed filesystem. The problem is that the USPTO is allowing obvious stuff to be patented. They even admit as such - unfortunately I can't find the article - but I remember reading the USPTO saying "only 5% of patents they grant are what they call pioneer patents" (in other words, something really new and worthy of patent protection). The reform needs to be that only these "pioneer patent" applications actually get granted and the rest thrown out.

Re:Really!

[i]

Yes; maybe; and the whole summary is stupid. From claim one of the patent; the very first paragraph:

having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers

So; where the mainframes of the 70s had single consolidated disks stores this is talking about doing this on a distributed filesystem. The area of application is indeed new completely opposite to the claim of the summary.

Patents are not supposed to control what you do; instead they control how you do it. Since the way that Google is claiming to do this is by going around comparing the timestamps on a bunch of different distributed chunks of a file, this is something that no mainframe of the 70s is likely to have had to to so it may even be a new way to automatically delete temporary files. I wish people would begin to understand this and commenters would point it out every time. I wonder if this isn't a bunch of patent lawyers trying to make us look silly.

I can't see a significant difference. In both cases they are using a metadata *expire* date. The only difference is that the files in question are distributed and the format of the metadata. In both cases they are comparing the metadata date with current date for files.

Re:Really!

[tambo]

> This is supposed to be new....

If by "that" you mean the invention described in the title - "Automatic Deletion of Temporary Files" - then, no.

Patent titles are as meaningful as book titles: you wouldn't assume that two books entitled "Pirate Adventure" relate the same story, right? It's the same with patents: a completely new type of automobile engine might have the title, "Automobile Engine."

If by "that" you mean the invention described in the independent claim, which is this -

1. A computer-implemented method comprising: selecting a file having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers, wherein each chunk has a modification time indicating when the chunk was last modified, and wherein at least two of the modification times are different; identifying a user profile associated with the file; determining a memory space storage quota usage for the user profile; deriving a file time to live for the file from the path name; determining a weighted file time to live for the file by reducing the file time to live by an offset, where the offset is determined by multiplying the file time to live by a percentage of memory space storage quota used by the user profile; selecting a latest modification time from the modification times of the plurality of chunks; determining that an elapsed time based on the latest modification time is equal to or exceeds the weighted file time to live; and deleting all of the chunks of the file responsive to the determining.

...then presenting that invention as "new" seems legitimate. There are several details in here - dealing with a temporary file as chunks across several file stores, each chunk having a different modification time; and determining the "weighted file time to live" based on the last modification time and the percentage of consumed file quota - that seem completely new.

The author of this Slashdot post appears to have glanced at the claims, reached the conclusion that "basically, it's about deleting temporary files," and posted this rant about how the patent office granted a patent for "deleting temporary files," inspiring yet another wave of diatribes about the patent office based on a faulty assumption. Not surprising - this kind of tilting at windmills, based on factually incorrect interpretations of patents, is a daily occurrence here.

Re:Really!

[tambo]

> If you had a distributed file which kept a timestamp on each of several separate chunks, how would you go about deciding when to automatically delete it? My guess is that the solution you would come up with quickly is basically the one in the patent.

Well, there are several ways you could deal with that problem. Here are some of them:

• * Deal with each chunk separately. Just let each machine decide when to delete its chunk.
• * Consider all of the chunks to have been modified as of the latest modification date on all of them. Sort all of the temp filed by modification date, and cull the oldest ones first.
• * Consider all of the chunks to have been modified as of the earliest modification date on all of them. Sort all of the temp filed by modification date, and cull the oldest ones first.
• * Consider all of the chunks to have been modified as of the average of the modification dates. Sort all of the temp filed by modification date, and cull the oldest ones first.
• * Consider all of the chunks to have been modified as of the file date listed in the shared filename. Update the modification dates accordingly, and then let each machine deal with its chunk independently of the others.
• * Consider all of the chunks to have been modified as of the file date listed in the shared filename. Update the modification dates accordingly, sort all of the time files by modification date, and cull the oldest ones first.

...etc. There are many, many variations on this technique that you might imagine. The one described in this patent is different from all of them:

deriving a file time to live for the file from the path name; determining a weighted file time to live for the file by reducing the file time to live by an offset, where the offset is determined by multiplying the file time to live by a percentage of memory space storage quota used by the user profile; selecting a latest modification time from the modification times of the plurality of chunks;...

...which is why the patent was issued.

Re:Really!

[Anonymous Coward]

Yes; maybe; and the whole summary is stupid. From claim one of the patent; the very first paragraph:

having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers

So; where the mainframes of the 70s had single consolidated disks stores this is talking about doing this on a distributed filesystem. The area of application is indeed new completely opposite to the claim of the summary.

BZZZT! You also don't know your history. HSM and friends were not restricted to contiguous blocks on a single system.

The idea's so old it was probably first implemented on an 8-bit SEL drum memory behemoth programmed with front panel switches. I've seen it re-invented a hundred times. Now GTFO my lawn you young whippersnapper!

Re:Really!

[hairyfeet]

Then the USPTO would lose the money they get paid for re-examinations which IIRC is the source of a good portion of their operating budget. This of course only encourages the examiners to pass bad patents as they'll be more likely to be called into question and thus be re-examined.

Re:Really!

[operagost]

I think OpenVMS has had file expiration dates since V2 at least... whenever the ODS-2 file system was created.

Re:Really!

[operagost]

I disagree. Just as half the patents trolls now take something in common use and append "on the internet" to it, this is taking something ancient and appending "on a distributed file system" to it.

Re:Really!

[SuricouRaven]

"deriving a file time to live for the file from the path name"

Sounds like an ugly hack to avoid modifying software to call the 'set expiration time' function.

Re:Really!

[rtfa-troll]

No; I agree. You are right in a sense. The sense is that, when you know that you have a distributed file system in which there are timestamps distributed to different chunks, every single programmer you can find will realises that you have to compare your expiry period against all of the distributed timestamps (at least until you find a newer one). However, until you have the actual technical details of some distributed filesystem you can't know exactly what you have to do to deal with it. That's the trick the judges use to pretend this isn't obvious.

Re:Really!

[Anonymous Coward]

I seriously doubt you implemented anything close to what is described in the patent. You may have implemented something to remove temp files from a distributed system after a certain amount of time or maybe even after a certain amount of time without being accessed. That is not what is described in the patent.

Re:Really!

[tambo]

> Sounds like an ugly hack to avoid modifying software to call the 'set expiration time' function.

Often, what looks like an "ugly hack" turns out to be an elegant, lovely solution for a peculiar scenario.

In this case, the solution doesn't require modifying software, the file system, the network protocol, or other metadata. That might make it more appealing than the "obvious" solutions to the problem.

Re:Really!

[rtfa-troll]

> If you had a distributed file which kept a timestamp on each of several separate chunks, how would you go about deciding when to automatically delete it? My guess is that the solution you would come up with quickly is basically the one in the patent.

Well, there are several ways you could deal with that problem. Here are some of them:

• * Deal with each chunk separately. Just let each machine decide when to delete its chunk.

doesn't solve the right problem; a file which has been recently used can be lost when one of it's

• * Consider all of the chunks to have been modified as of the latest modification date on all of them. Sort all of the temp filed by modification date, and cull the oldest ones first.

this is (one of) the solution(s) given in

• * Consider all of the chunks to have been modified as of the earliest modification date on all of them. Sort all of the temp filed by modification date, and cull the oldest ones first.

doesn't solve the right problem; a file which has been recently used can be lost when one of it's chunks is not updated

• * Consider all of the chunks to have been modified as of the average of the modification dates. Sort all of the temp filed by modification date, and cull the oldest ones first.

doesn't solve the right problem; a file which has been recently used can be lost when one of it's chunks is updated but the others all become old

• * Consider all of the chunks to have been modified as of the file date listed in the shared filename. Update the modification dates accordingly, and then let each machine deal with its chunk independently of the others.

doesn't solve the right problem; you can no longer refer to a file by a fixed name

• * Consider all of the chunks to have been modified as of the file date listed in the shared filename. Update the modification dates accordingly, sort all of the time files by modification date, and cull the oldest ones first.

as previous answer

...etc. There are many, many variations on this technique that you might imagine. The one described in this patent is different from all of them:

deriving a file time to live for the file from the path name; determining a weighted file time to live for the file by reducing the file time to live by an offset, where the offset is determined by multiplying the file time to live by a percentage of memory space storage quota used by the user profile; selecting a latest modification time from the modification times of the plurality of chunks;...

...which is why the patent was issued.

There are lots of things which are different from the patent, for example going around the block. However, that doesn't mean that the patent isn't an obvious solution to the given problem when added to the prior art. For example; let's say I want to get the number 105353284. Probably nobody has explicitly published a method of getting it by adding 105353175 to 109. However that doesn't mean I get to have a patent on it. The fact that addition exists overall is prior art. Just like the fact that comparing timestamps and looking through lists exists is should be prior art to this patent.

Re:Really!

[chrismcb]

Surely this is novel. I mean who would have thunk of automatically deleting temp files (well based on the junk in my temp directory, not many people) AND figure out a way to solve it? It must have taken several Phds and a plethora of light bulbs to figure this one out.

Re:Really!

[chrismcb]

Patents are not supposed to control what you do; instead they control how you do it. Since the way that Google is claiming to do this is by going around comparing the timestamps on a bunch of different distributed chunks of a file, this is something that no mainframe of the 70s is likely to have had to to so it may even be a new way to automatically delete temporary files. I wish people would begin to understand this and commenters would point it out every time. I wonder if this isn't a bunch of patent lawyers trying to make us look silly.

So what you are saying is one could add the words "distributed file system" to an old patent, and get a new patent? And this patent really isn't about DFS, as it just talks about finding the most recent modified time stamp of the distributed file. No the main portion seems to be creating a ratio between the time to live and the amount of disk space left.
I don't think that adding "on the internet", "on a tablet", "on a DFS", "in the cloud" is enough to create a new patent. Now if the system to allow you to use the new technology is novel, then sure. But getting the latest timestamp of a collection of timestamps? No that shouldn't be patentable in any way shape or form.

Re:Really!

[chrismcb]

The 'plurality' of chunks is irrelevant to the matter of expiring old data. What matters is that you have some sort of metadata telling the system when a blob of data that might resemble a file is to be deleted.

I believe that is what the summary claims, and why the summary believes it is similar to the older patent. But really its more about modifying the "metadata" based on the amount of space left. Again no big stretch, and nothing patent worthy about it.

Re:Really!

[SuricouRaven]

Until your log filename turns into access_log_0213_expire.-.02132020

hmmm

[Anonymous Coward]

I wonder if google is going to chase me. I use that exact method for log file expiration in a program I wrote back in 1998 for scanning configurations across servers, From memory I also got the code for doing that from someone elses web site that had posted the sample. the log files are written to dated folders according to how long I needed to keep a record for of the specific scan being executed.

Re:hmmm

[viperidaenz]

Did you also inspect the quota for the user owning the file to determine if you should delete it? Were the files also stored in a distributed file system, with chunks of the file on separate systems?

Every single claim in that patent mention both of those things.

The naming of files is an example of a part of a claim. To infringe on a patent you need to infringe on at least one entire claim.

Re:hmmm

[Anonymous Coward]

Oh please. This is hardly worthy of a patent.

Re:hmmm

[Anonymous Coward]

yes it was stored on a distributed file system (still is), yes quota was also checked as that was the basis for cleanup, clean-ups would use the age to determine how best to reclaim space. but no to the chunks on separate systems. Still it is hardly an original idea worthy of a patent.

Re:hmmm

[shutdown -p now]

Did you also inspect the quota for the user owning the file to determine if you should delete it?

Obvious enough that it's not patent-worthy.

Were the files also stored in a distributed file system, with chunks of the file on separate systems?

Doesn't seem to have any relation to the basic principle being patented - unless you're claiming that, after someone patented the wheel, I can come in and patent the use of wheel specifically on paved surfaces.

Re:hmmm

[viperidaenz]

Yes it does, every single claim in the patent mentions it or expands on another claim that does. Along with other details in every claim

Re:hmmm

[shutdown -p now]

What I meant is that I don't see how the fact that the temp. files being deleted are on a distributed file system is relevant to the concept of cleaning them up?

Re:hmmm

[Genda]

Oh! Oh! I know! Just add that it has to be performed on Tuesday at 8:43 AM by a trained seal named Fernando! Yeah... that'll do it

Re:hmmm

[rtfa-troll]

What I meant is that I don't see how the fact that the temp. files being deleted are on a distributed file system is relevant to the concept of cleaning them up?

It's because there's no concept of the inode where I can go and see a single timestamp for the age of the file. Instead I have a timestamp on each of the chunks which make up the file. This adds the following to claim one

selecting a latest modification time from the modification times of the plurality of chunks

Which legally, according to East Texas and the Federal circuit, but probably not according to the more recent decisions of the supreme court of the USA, is sufficient to make this patentable material. Note that it doesn't tell you some magic new fast algorithm for doing this. There is nothing valuable in this patent. However, if Google didn't patent it Microsoft would and would then sue them in East Texas and have the court ignore the fact that Google had published it as prior art. They have little choice but to do stupid patents as long as the stupidity of software patents continues.

Re:hmmm

[sFurbo]

Doesn't seem to have any relation to the basic principle being patented - unless you're claiming that, after someone patented the wheel, I can come in and patent the use of wheel specifically on paved surfaces.

You can. Or, to use the example I have been taught, if you find a new use of a substance, you can patent that use, regardless of any patent one the substance. Your use have to be novel and non-obvious, of course, and you might not get much use of your patent for as long as the substance is under patent by someone else. They can't use it in the way you patented, though.

I apologize in advance if this is not at all relevant to the discussion, I have really not have enough coffee today.

Re:hmmm

[osu-neko]

Your use have to be novel and non-obvious...

Not anymore. Might not stand up in court if it isn't, but you'll be able to make a bunch of money patent trolling a lot of individuals and small companies before anyone actually challenges it in court. If you're carefully choosing your targets, you have a good chance of avoiding that even happening.

Re:hmmm

[Jafafa Hots]

I just invented a real-world application of the same thing Google is patenting here.
Real world sorting of items for disposal.

The way it works is there is a marker on the piece of trash. A triangle with a number in it. I will use my brain program to check that marker and decide whether the article needs to be put in the recycle bin or in the regular trash.

I'm gonna be rich!

Re:hmmm

[Theaetetus]

Did you also inspect the quota for the user owning the file to determine if you should delete it?

Obvious enough that it's not patent-worthy.

In hindsight, sure. But to show it was obvious at the time the application was filed, you need to show that that element exists in one or more prior art references.

Think of it like convicting someone for murder. You can't just look at them and say "they look guilty", you have to provide evidence to support that conclusion. Saying the claim is "obvious" is a conclusion, too, that has to be supported by evidence.

Re:hmmm

[shutdown -p now]

I'm not a patent examiner, so I can use my common sense and industry experience to make this conclusion. It's obvious that it's obvious.

Also, this is not a court of law, and we're not debating whether we should stick Google into prison for murder; quite the opposite, we're debating giving them the privilege to prevent other people from using something. If anything, they should be the ones to prove that everything is in order, including that it's not obvious, if anyone claims otherwise.

Re:hmmm

[Theaetetus]

I'm not a patent examiner, so I can use my common sense and industry experience to make this conclusion. It's obvious that it's obvious.

I disagree. But that aside, my point is that the PTO and the courts are bound by the constitution and the requirements of due process, so they can't merely decide a patent is invalid without evidence.

Also, this is not a court of law, and we're not debating whether we should stick Google into prison for murder; quite the opposite, we're debating giving them the privilege to prevent other people from using something. If anything, they should be the ones to prove that everything is in order, including that it's not obvious, if anyone claims otherwise.

How do you prove that something wasn't obvious? Prove that no one, anywhere in the world, in the entirety of history, had that idea before? Should we conduct surveys using our time machine?

There's a reason the burden of proof is on the patent examiner, just like there's a reason the burden of proof is on the prosecution. Justice, as well as logic, demands it.

Re:hmmm

[shutdown -p now]

How do you prove that something wasn't obvious? Prove that no one, anywhere in the world, in the entirety of history, had that idea before? Should we conduct surveys using our time machine?

It depends on the threshold. I'm not asking for "beyond reasonable doubt here", but rather something more akin to "preponderance of evidence". This is something that could be solved by a panel of industry experts, for example.

There's a reason the burden of proof is on the patent examiner, just like there's a reason the burden of proof is on the prosecution.

As noted, the difference is night and day. In patent cases, the patent owner is really a defender - rather, we all (the society) are, because he is seeking to limit our freedom for the sake of his profit. I think it's a good idea to make it so that any such limit has to be proven to be reasonable, as opposed to us having to prove every time that the limit is unreasonable. After all, we have been practicing the latter for a long time, and look what a mess the patent system has become today.

Re:hmmm

[Theaetetus]

How do you prove that something wasn't obvious? Prove that no one, anywhere in the world, in the entirety of history, had that idea before? Should we conduct surveys using our time machine?

It depends on the threshold. I'm not asking for "beyond reasonable doubt here", but rather something more akin to "preponderance of evidence". This is something that could be solved by a panel of industry experts, for example.

The problem there is that the application sits in a backlog for a couple years before it gets examined, and in that time, the inventor may have released product, given lectures, shown around white papers, taught classes, etc. explaining the invention. It may be common knowledge to those experts and obvious years later - just like you or I could probably sketch a model of an internal combustion engine on a napkin right now - but that doesn't mean that it was when the application was filed. You need a panel of "experts" that aren't actually experts now, but have been in isolation from the state of the art for several years.

There's a reason the burden of proof is on the patent examiner, just like there's a reason the burden of proof is on the prosecution.

As noted, the difference is night and day. In patent cases, the patent owner is really a defender - rather, we all (the society) are, because he is seeking to limit our freedom for the sake of his profit. I think it's a good idea to make it so that any such limit has to be proven to be reasonable, as opposed to us having to prove every time that the limit is unreasonable. After all, we have been practicing the latter for a long time, and look what a mess the patent system has become today.

What mess? And it's actually pretty easy to show something is obvious - you just have to find each of the elements in the prior art, even if they're not in the same reference. Like, if A, B, and C are known, then a patent claiming A+B+C is obvious. But if no one's ever heard of C before, then how can that patent be obvious?

Re:hmmm

[viperidaenz]

Do you look at all the other pieces of the paper and check the numbers of them too, to make sure you're looking at the latest number for that type of trash, where those other pieces are in different locations? Is that number also translatable to a time, when up until that time, or a later date depending if there is still ample capacity, you'll keep the trash around because you don't really need to throw it out and someone else may need to still use it?

None of you people have actually read the patent have you?
TFS picks out an example of a variation on a small part of a claim, that isn't itself the idea being patented to cause an uproar.
This is part of their filesystem, no other file systems do it the same way.

Re:hmmm

[Anonymous Coward]

It looks to me (at first glance) to be something intended for enabling the individual nodes to be able to determine when & whether they ought to delete their "file chunks," rather than having a process that has to traverse a remarkably long directory list. Just a guess... because this sort of maintenance becomes a much more interesting problem to solve if one thinks about just how large "cloud storage" files & filesystems could get... imagine instead a UNIX-like system with visibility over a distributed filesystem hosting files for such a *large* number of active customers, that even with the best cache available, "ls" would just start a never-ending process that never completes.

The real problem

[WindBourne]

is that the USPTO inspectors are mostly foreign born and do not have experience with this older gear and innovations. Worse, many are foreign educated. They are now doing time in USPTO to learn english to get real jobs.

Re:The real problem

[cheater512]

Cut the foreign born crap (Aussie here). Just say they are incompetent and leave it at that. Its more accurate that way.

Re:The real problem

[Anonymous Coward]

Really. You know this how? A know a lot of people born in the US who can barely speak English.

Re:The real problem

[Anonymous Coward]

May be USPTO should hire older IT guys displaced by H1B as patent examiners?

Re:The real problem

[Anonymous Coward]

Because Spanish is their first language?

Re:The real problem

[Anonymous Coward]

North Americans are so cute with their "excepcionalism" syndrome.

Re:The real problem

[Anonymous Coward]

I highly doubt that that's the "real problem", given that you are legally required to be a U.S. citizen to become a USPTO inspector.

http://careers.uspto.gov/Pages/PEPositions/fitcheck.aspx

Re:The real problem

[Anonymous Coward]

I do too. They're all niggers.

Re:The real problem

[WindBourne]

No, it is NOT. The fact is, that even Aussies from the 70's had western goods. Others who were outside of the western sphere did not get as many goods, nor were participating in development of them. For example, how many innovative Chinese and Soviet items did you have from when you were a kid? I will bet NONE. Yet, both had them.

Re:The real problem

[WindBourne]

Your point? I deal regularly with foreign born. My wife is foreign-born. My in-laws speak tamil and many of my ex-gf's families only spoke spanish. So, really, what is your point? Do you have anything intelligent?

Re:The real problem

[WindBourne]

Becoming a US citizen does NOT mean that you have command of the english language. If you look at the CVs of the inspectors, you will find that many of them have engineering degrees from China or India. Basically, they do NOT have the exposure to the same innovations that somebody raised here has.

Re:The real problem

[WindBourne]

To be honest, that is not a bad idea. BUT, keep in mind that only solves IT. You have engineering as well. And in 20 years or so, you will want to have a number of asian-born inspectors. IOW, at this time, they are too early.

Re:The real problem

[WindBourne]

And yet, we have so many idiots with their everything-is-racist-or-against-me syndrome, that they are incapable of looking at issues realistically. Hell, you have to hide who you are.

Re:The real problem

[Jmc23]

Please don't lump us canadians and mexicans in with the USians ...and some quebecers.

Re:The real problem

[Anonymous Coward]

So, you like foreign pussy. Your point?

Re:The real problem

[cheater512]

Oh sorry I didn't realise everyone had a 70s mainframe back then. My mistake.

Heck did you know those mainframes had that feature? I didn't.

Re:The real problem

[Daniel Dvorkin]

Hell, you have to hide who you are.

Yes, I'm sure your parents were so proud when they looked down at you squalling in your crib and said, "We'll call him WindBourne, that's such a wonderful name!"

Re:The real problem

[WindBourne]

Well, first off Dan, you already know who I am. We have emailed plenty before. Secondly, I use the same item here. There is only ONE person as Windbourne on /.. OTOH, there are plenty ACs here.

Re:The real problem

[WindBourne]

Actually, I DID. Why? Because I worked on mainframes as well as early unix, DOS, Xenix, etc. However, how many Chinese or Indians from the 70's, or even 80's, worked on IBM mainframes? Probably none that are working in the USPTO. BUT, I will bet that there are PLENTY of schools around Australia that have/had mainframes, PDPs, VAXs, etc. In fact, I know that even back in the 90's, Colorado State's main programming was still done on mainframes. There were plenty of other schools around the western world that did the same thing. The reason is that it costs money to change.

Re:The real problem

[Daniel Dvorkin]

Okay, I'm puzzled. Do we know each other outside of /.?

All I'm saying is that it seems to me that posting under a screen name isn't a whole lot different from posting as AC, in terms of willingness to reveal your identity and stand by your words. You may have a certain amount of reputation and karma to gain or lose, but that's about it. Now, when I signed on to /. using my real name, it wasn't a deliberate act of courage or anything like that--I just didn't even think about coming up with a screen name--but in retrospect, I do take a certain amount of pride in it. My opinions are out there with my name attached to them for all the world to see, and I think I write more carefully and thoughtfully than I would under a pseudonym.

Re:The real problem

[Zontar The Mindless]

As someone who works with Indian engineers daily, who lives with (and is about to marry) a Chinese engineer, and who is himself "foreign born"... You are so full of racist crap, I'd be afraid to kick you out my door for fear of ruining the carpet.

Re:The real problem

[WindBourne]

Okay, I'm puzzled. Do we know each other outside of /.?

At one time, yes. I got my degree from Colorado State in Microbio/genetic engineering, worked at CDC, also worked as EMT. Then got into software engineering including at Metpath and CU-med. LONG ago.
in addition, we have emailed before.

Re:The real problem

[WindBourne]

One last thing, regardless of wether I operate under a pseudonym or my real name makes no difference to me. I express the same opinion all around. I have never had a friend call me racists or think that I was. The fact is, that I use a pseudonym because I was well aware that spammers were coming. My initial approach was to avoid them by not registering. Now, I use one pseudonym and try to keep it quiet (not like it is that hard, but I hate spammers). In addition, now, I have a sociopathic ex, so, I stick with this.

Re:The real problem

[WindBourne]

I am MARRIED to an Indian software engineer. I have worked with plenty of indian / chinese software engineers. There is NOTHING racists in what I have said. India was close to USSR until around 2000 (basically, once USSR broke apart, then they fell more into the western sphere). China has ALWAYS worked to keep western influences and goods out ( most likely due to the opium wars).

The only racist individual is yourself. The reason I say so is that you do not see the issue because you are afraid of being called a racist, rather than looking at something. To add to that, I am guessing that you have ZERO experience with USPTO. And yes, I have experience dealing with patents.
Enave, Yermai poolu poite chhapeda, naye!

Re:The real problem

[Redmancometh]

I call bs mr unicorn.

Re:The real problem

[Genda]

What??? You didn't have a Univac in the coat closet? The local Jr. College was dumping there's... not only a great machine... a fine space heater!

Re:The real problem

[hjf]

Oh, you americans are soooo innovative and creative, and special.

So much, you had to steal Von Braun from the nazis to make your useless rockets work.
So much, Intel had to recruit former soviet engineers that came up with the idea of a superscalar processor.
So much, you had to get a guy from Argentina to help you figure out coronary bypass.
So much, your best movies (Harry Potter series, LOTR, The dark knight...) are written and directed by foreigners (british! new zealanders!)
So much, THAT FUCKING GOOGLE WAS CO-FOUNDED BY A RUSSIAN.

I could keep adding examples all night, but it won't matter. You are just stupid. You are a redneck with a degree. You have to be a complete tool to think "they do NOT have the exposure to the same innovations that somebody raised here has". How can you be just oh-so-fucking-stupid? Have you ever even set foot outside the US? Did you know down here we also have things like, cars, and roads, and vaccines and cell phones?

Asshole.

Re:The real problem

[dbIII]

Australia that have/had mainframes, PDP

There was even a movie called "The Dish" that featured one. Other bits from that movie were fictional (transmissions to and from Apollo 11 went off without a hitch) but when the studio went looking for a PDP11 for a prop they found the original from Parkes Observatory and found it could still be powered up. There's another which was used in a final year electrical engineering project five or so years back where the students interfaced a few gigabytes of unwanted 72 pin memory to a PDP(10 or 11?) and worked out some way for that old machine to address all of it.

Re: The real problem

[Anonymous Coward]

How much did Facebook pay you for that post?

Really, it takes some skill to take someone who values their privacy and make them look like a coward.

Re:The real problem

[silentcoder]

Sshhh don't tell him but third world South Africa was running it's social welfare program with a database on a mainframe back in the 1960s. Punch-cards and all.

I actually met the guy who was the chief operator/programmer on it once (in the late 90's), he declared in conversation that "I wrote code to manage a database of millions of entries on a computer with 64Kb of RAM that filled an entire floor in our building. The commodore 64 had the same memory and three times the CPU power in 1980 and you could carry it in a briefcase. That was good, the downside is: programmers nowadays don't know how to write efficient software anymore".

Many of the great computer companies of the past were NOT American, and even the ones that were exported widely. I had a comodore 64 in the early 1980's and I used it's manual to learn to program a few years later using Microsoft BASIC on a 286. A few years after that I was learning Turbo Pascal - and this was all before I went to high-school.

We in the rest of the world did in fact have access to many of the same technologies the US did, and when we did not - we generally DID know about it, and of course South Africa was probably the LEAST exposed to such technologies because until 1989 we were under export sanctions so companies in the US weren't ALLOWED to sell us stuff.

Not that this stopped most of them - all they did was create a wholy-owned company registered INSIDE South Africa that manufactured the goods locally. If anything sanctions created a great deal of employment for South Africans and actually postponed the end of apartheid. It had some weird consequences too: Pepsi refused to use the loophole (claiming they supported the anti-appartheid movement), Coca-Cola had no such qualms and built a bottling plant in South Africa. As a result: to this day, Coca-Cola remains the preferred soft drink here (even among black people) and pepsi hardly sells in the few stores that bother to stock it at all because two generations of South Africans grew up in a world where Pepsi didn't exist but Coca-Cola did.

Re:The real problem

[Big Hairy Ian]

Ahh bigotry I knew there was something missing from slashdot this morning! I will agree that the staff of the USPTO do seem moronic but then America has morons just like everyone else!

Re:The real problem

[WindBourne]

It amazes me how intellectually downwards /. has become. When you do not like it, then you scream racism, bigotry, etc.
YET, the fact is, that many of those that were either under USSR or Chinese spheres while growing up, had access to different technologies, with different education. MOST of the Comp.Sci and electronics that are used today, were developed in the west. Yet, you seem to think that somebody who had ZERO access to this, would know it.
So, let me explain it to you real slow and simple for you. The examiners already have access to previous patents via a simple DB search. Likewise, any competent examiners can read a patent and decide it is done up decently enough to be considered a patent. Likewise, any competent examiner will know such things as perpetual motion machines do not exist unless somebody can produce a working example of these and even then, look closely.

SO, where do we run into issues with PRIOR ART? It is when it was developed long ago, BUT WAS NOT PATENTED, OR IS NOT TAUGHT IN THE BOOKS. Therefore, you had to have worked in the OLDER languages, OSs, equipment, hardware, etc. to know about it. BUT, if you come from a nation that did not have such, then you have NO PRIOR EXPERIENCE OR FOOTING on which to know this. Now, that means that not only Chinese and Indians, BUT the old Soviet union itself, loaded with white ppl, will have an extreme disadvantage in determining if a patent has USED PRIOR ART. Of course, very few of the examiners are from old USSR, so this is a none issue.

One simple way to rectify these issues is for USPTO to run all of the patents that were approved by current examiners through a small group of older engineers, coders, who will be looking for ideas that existed before. This will not catch them all, BUT, they will catch many of them.

Hopefully, the above will help you to realize that the ONLY racism here is YOURS (and the other trolls) in ASSUMING that this was about racism/bigotry, when it was about nothing BUT FACTS. Yet, I have a feeling that you not give a single thought to this and will simply deny it. You (and a few others here) are like those who oppose Climate Change or scream that god created us. You simply chose to avoid facts while screaming that the other person is prejudiced in some way.

Re:The real problem

[WindBourne]

Hmmm. Where exactly did Von Braun get his rockets from? Oh yeah, Godard, an American.
Oh yes, Seymour cray who developed superscalar in the 60s, was definitely Russian. Or the fact that Intel had superscalars PRIOR to the fall of the USSR really proves your points.
Corinary Bypass was an ex-german who did some of the work in SOuth Africa and was able to do it when he hooked up with an American. The actual work was done by the American, not the ex-german. Oh, as to Argentina, please look at the last section of the history. You will notice the argentinian that you speak of, did his work 7 years later. Here is a comment from one of the docs who was there, but hey, do not let what those who were there influence YOUR idea of facts.
WHO in their right mind thinks that Harry potter was American or LOTR was American? Harry is 100% English filmed and LOTR was filmed in New Zealand. I think that ANYBODY with any intelligence KNOWS both of these.
And Sergey MOVED to the USA when he was 6 years of age. IOW, he grew up as a first generation American. My wife moved here at age 10. And I can tell you that she is neither British nor Indian except in birth and ancestors, respectively.

And yes, I have been out of USA and multiple continents. I would guess far more than you have been to other continents.
You could keep adding stuff, but the fact is, that you were wrong on every single thing that you wrote. Sadly, your racism and lack of intelligence is getting in your way.

Re:The real problem

[WindBourne]

Exactly. That is because Australia is a Western nation. BUT, you will only find mainframes, PDPs, and Vaxes in special locations of USSR and China that were about trying to copy what we were up to.

Re:The real problem

[WindBourne]

Shhh. Be very quiet them. But, here is a thought for you. How many south Africans WROTE the MVS system? Or how about VM? Writing a program that makes use of OS, DBs and libraries generally does not involve that much work. And those that wrote on punch cards with fortran, cobol, JCL, RPG, etc. did NOT develop elaborate algorithms. The algorithms were discovered in languages like Assembler (including Mix, but I suspect that you do not know that one; z-80s, 8086, etc), PL/1, etc. Heck, on the series/1 we did elaborate algos, but that was all with assembler.

Re:The real problem

[Anonymous Coward]

And hence a bushman travels to the end of the earth to rid his tribe of the evil thing

Re:The real problem

[Anonymous Coward]

Wow the op was obviously racist sorry but if you can't see it he'll probably acuse you of being foreign because of your lack of grasp of english!

Re:The real problem

[silentcoder]

Actually - quite a few of them were involved with those systems (though mostly they were expats living the USA by then).
By the time the free software movement started taking hold quite a few of the pioneers were from South Africa. One of the first was Paul Sheer - who wrote mcedit (the text editor bundled with midnight commander) for example.
Another South African born engineer Theo De Raadt wrote OpenBSD.

Re:The real problem

[WindBourne]

Those were from the 90's. The mainframe era was from the 60's through 80's. Totally different items. All of the dos items were fairly trivial apps that simply copied items from the mainframes.

And the openBSD was NOT written by De Raadt. He forked it and continued to modify it.

Re:The real problem

[silentcoder]

Yes, I know that, I gave free software examples knowing they were later because I actually know their names.
That doesn't mean there weren't expat engineers working for IBM and DEC in the age of PDPs, I know for a fact there was.

Re:The real problem

[WindBourne]

Oh, I am sure that some expats were here. However, it was nothing like it is today. Back in the 60-80,s the vast majority of computer software was done by the west and more importantly, by the locals. That is why I continue to say that when looking for unpatented software, the right way is to pick amongst the older ppl that did this work. Likewise, in the future, more engineering and software patent examiners will have to be foreigners. But, at this time, the right group would be the westerners that did this work.

Re:The real problem

[hjf]

Dude, I don't give a fuck, the people that ACTUALLY MADE THINGS WORK weren't american. Please, go fuck yourself with a cactus, please!

Wait there's more

[Anonymous Coward]

Google Labs is supposedly working on a next-gen programming development environment that allows source code statements to be physically manipulated like a deck of cards.

Re:Wait there's more

[Anonymous Coward]

You mean they're making a remake of Solitaire?

Re:Wait there's more

[93 Escort Wagon]

Kids today...

Re:Wait there's more

[sycodon]

That's pretty fucking funny and I'm not even drunk.

Even if you were being sarcastic.

If you're not....PLEEEASE tell us!

Re:Wait there's more

[fatphil]

If that's supposed to be funny - shouldn't there be a punch line?

I'll get me coat...

Re:Wait there's more

[Anonymous Coward]

And, as an extension, I will patent putting a "sequence number", which is a monotonically increasing number, in specific columns (say 73-80) so that the data can be reordered should the "cards" become "scrambled" by some inadvertent process akin to "dropping the deck".

Re:Wait there's more

[jfdavis668]

For those who do not get the joke, he is referring to the 70's when programs were written on IBM punch cards, and you could change the program by physically moving them. Punch cards are actual paper with holes punched in them to be machine readable. Sorry about being so obvious.

Re:Wait there's more

[Anonymous Coward]

Like hypercard ?

Re:Wait there's more

[Anonymous Coward]

Right? All they think of are those violent games. They should ban them before some kid goes crazy and starts shooting people or something.

Re:Wait there's more

[Anonymous Coward]

You mean they're making a remake of Solitaire?

Original poster is saying that Google Labs is going to patent the long-forgotten punch-card of the 1960's and '70's

Re:Wait there's more

[previewlounge]

oh dear that is hilarious! totally made my evening.

big deal

[dickens]

It's not like Microsoft was ever going to be interested in that anyway. They must get cents back from the disk manufacturers for perpetuating their ever-growing temp folders.

Re:big deal

[hedronist]

ROFLCOPTER! Being the Geezer Geek® in our family/neighborhood I get the calls for ... almost everything. I had one woman (who shall remain nameless (except she's my older sister)) who complained of all sorts of horrible things happening. Turns out it was an out-of-disk and there was a huge number of ~whatever files all over her filesystem. Cleaning them all up gave her about 40% free disk.

Re:big deal

[Anonymous Coward]

It's fine to name the woman. She has nothing to be embarrassed about. It is the operating system company that should hide in shame.

really?

[lkernan]

I don't know whats worse, Google applied for this, or the USPTO approved it.

Re:really?

[edibobb]

It's much worse that the USPTO approves trash like this. As long as they do, companies, trolls, and other profit seekers will take financial advantage of the stupidity. Meanwhile, the more patents the USPTO approves, the larger their budget becomes.

Re:really?

[Capt.Albatross]

I don't know whats worse, Google applied for this, or the USPTO approved it.

I am no particular fan of Google, but if the USPTO is approving this sort of thing, Google (and every other business) has to worry that some troll will land a patent on some basic part of their everyday operations, and if you can afford it, one defense is to attempt to patent everything that you do and use. They may have been as surprised as we are that it was approved.

For the USPTO this is a wonderful business model: do a crappy job and increase the demand for your services. Another recent example of this business model in practice was the way S&P rated mortgage-backed securities.

Re:really?

[Anonymous Coward]

grant 'em all and let the courts sort 'em out

Re:really?

[Big Hairy Ian]

Can I suggest somebody tries to Patent a "Negative IQ" that way the USPTO is an automatic demonstration of prior art!!

Re:really?

[chrismcb]

The problem is, everyone else is doing it. So you pretty much have to. You can't really blame a company for playing, in fact you can blame them for not playing, as they aren't doing everything they can to protect their assests. AND they have to leverage in the patent wars.
You can blame USPTO and Congress for this mess.

No issue here, Read the Patent!

[dajjhman]

If you actually read the patent, it is specifically for a similar method, but designed for Distributed File Systems. This is different from just a single file being names a certain way. It is an algorithm based on the location of other related files, each different file's modified and Time to Live (TTL) dates, and the factors determined by the, keywords here, plurality of servers. If they tried to patent a regular temporary file that would be different, but this is a distributed system specifically for a file that is distributed in different parts on different systems. If you still think this has been done before, I would love to see the source for that information and gladly would recant myself given that.

Re:No issue here, Read the Patent!

[dajjhman]

Forgot to add: Specifically read the Claims section.

Re:No issue here, Read the Patent!

[Anonymous Coward]

A distributed filesystem is still a filesystem. One doesn't get to patent filesystem features again (and especially not obvious ones) just because they're now "distributed".

Re:No issue here, Read the Patent!

[samkass]

Here is the crux of the first claim: "1. A computer-implemented method comprising: selecting a file having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers, wherein each chunk has a modification time indicating when the chunk was last modified, and wherein at least two of the modification times are different; identifying a user profile associated with the file; determining a memory space storage quota usage for the user profile; deriving a file time to live for the file from the path name; determining a weighted file time to live for the file by reducing the file time to live by an offset, where the offset is determined by multiplying the file time to live by a percentage of memory space storage quota used by the user profile; selecting a latest modification time from the modification times of the plurality of chunks; determining that an elapsed time based on the latest modification time is equal to or exceeds the weighted file time to live; and deleting all of the chunks of the file responsive to the determining."

Can we please have an end to the stupid articles where someone intentionally mis-interprets the abstract or even just the title of a patent and pretends it's some simple thing that's been done for decades to try to drum up anti-patent sentiment? There seems to be one a week or so.

Re:No issue here, Read the Patent!

[Anonymous Coward]

Embedding metadata in the filename? Unheard of!

Re:No issue here, Read the Patent!

[c0lo]

If you actually read the patent, it is specifically for a similar method, but designed for Distributed File Systems.

Ahhhh... that's good.

You see, I was scared shitless that we are still quibbling over patents granted with the only claimed difference over some old methods (patented or not) being "on a computer".
I see now how wrong I was: we stepped in the glorious era of the "in the cloud" claims.

Re:No issue here, Read the Patent!

[stevesliva]

Can we please have an end to the stupid articles where someone intentionally mis-interprets the abstract or even just the title of a patent and pretends it's some simple thing that's been done for decades to try to drum up anti-patent sentiment? There seems to be one a week or so.

Unlikely. Nonetheless, anti-patent sentiment is a good thing. Far too many people assume there's some sort of fairness or justice to the whole mess, and there isn't.

Re:No issue here, Read the Patent!

[Sarten-X]

Look, I know you're trying to be insightful and all, but this is Slashdot. We don't let silly facts get in the way of our unbridled hatred of all things Gub'mint!

It's a patent on storing files. Sure, it has some improvements that nobody's really used before that solve particular problems in a particular field, but I totally saw something similar sketched on the back of an envelope at my cousin's house in 1957, so this is blatantly obvious. I don't need to read the silly lawyer-speak claims to see that this patent is obviously just for storing files on a computer, in a distributed way. Heck, at my office building we have several rooms of filing cabinets, with several drawers each. If that isn't distributed, I don't know what is!

</PainfullyTrueSarcasm>

More seriously, yep. Knowing a thing or two about how Google stores data, this looks like a means to keep their GFS size needs under control. As each chunk is modified (but normally such chunks aren't able to be deleted), it can be given a whole new path, allowing a separate process to simply delete blocks that haven't been modified in a long time.

Re:No issue here, Read the Patent!

[Anonymous Coward]

One click "in the cloud" here I come!

Re:No issue here, Read the Patent!

[Improv]

It's still a dumb patent; a trivial weighting addition doesn't change this. I mean, seriously, that's less complicated than your average photoshop filter, and it's an obvious "innovation" that any engineer would think up if they were to be asked to implement file expiration on Google's platforms.

Re:No issue here, Read the Patent!

[Daniel Dvorkin]

Utility, novelty, and non-obviousness. This patent clearly fails at least one of these conditions.

Re:No issue here, Read the Patent!

[MightyMartian]

So if I add "over a network" to a claim that makes it patentable?

Re:No issue here, Read the Patent!

[Anonymous Coward]

Can we please have an end to the stupid articles where someone intentionally mis-interprets the abstract or even just the title of a patent and pretends it's some simple thing that's been done for decades to try to drum up anti-patent sentiment? There seems to be one a week or so.

This is slashdot, so I think the answer to that is obvious. But just in case it isn't obvious to you, No.

Re:No issue here, Read the Patent!

[Anonymous Coward]

It may not be the same as the prior art stated by the article summary I'll agree, BUT its still math. It's an algorithm. Math isn't patentable. The general argument against software patents remains and is exemplified by this one.

Re:No issue here, Read the Patent!

[1u3hr]

While the mechanism described isn't very original, Google has lawyers to cover their asses and patent it so that a few years down the trail when they've implemented it all through their cloud they can't get sued by Microsoft or some troll that went ahead and patented an equivalent method. Actually, if the PTO threw it out as unpatentable, Google would probably be just as happy, so they could use it without looking over their shoulders. .

Re:No issue here, Read the Patent!

[sycodon]

From trash is still trash.

Re:No issue here, Read the Patent!

[sycodon]

Software should not be patentable. Period.

Re:No issue here, Read the Patent!

[bill_mcgonigle]

One doesn't get to patent filesystem features again (and especially not obvious ones) just because they're now "distributed".

Bwahahahahha. That's the industry term for "only the suckers will wait to re-file every known filesystem patent with 'on a distributed filesystem' added to the end." Referencing prior art is for weaklings since weak patents can still be used to extract medium-sized license fees from end-users before they're invalidated. Set up a good corporation to take the fall for it.

Re:No issue here, Read the Patent!

[rastoboy29]

Wow.  Yeah.  It's an amazing innovation.  Totally patentable.

I mean, more than one server.  My mind is blown.

Re:No issue here, Read the Patent!

[Anonymous Coward]

Sounds like caching for a distributed system. Ain't nobody ever done that before...

Re:No issue here, Read the Patent!

[rtfa-troll]

Your definition of "obvious" is different from the Federal Circuit's. They mean something like "if you put a programmer in a room he would come up with this sponaneously" where you mean something like "if asked to solve a problem I could come up with this quickly and wouldn't need to look up a patent to do it". Your definition would be better and fit better with humans understanding of the word "obvious". It would also make more chance that patents would be valuable. Unfortunately, however, law doesn't work like that. Obvious is defined by the politicians, lawyers and judges. Very often the politicians deliberately twist the words so that the law seems acceptable when a layman reads it but they manage to screw the layman over for the interests of their big corporate buddies.

Re:No issue here, Read the Patent!

[Anonymous Coward]

If you actually read the patent, it is specifically for a similar method, but designed for Distributed File Systems. This is different from just a single file being names a certain way.
It is an algorithm based on the location of other related files, each different file's modified and Time to Live (TTL) dates, and the factors determined by the, keywords here, plurality of servers.

If they tried to patent a regular temporary file that would be different, but this is a distributed system specifically for a file that is distributed in different parts on different systems.

If you still think this has been done before, I would love to see the source for that information and gladly would recant myself given that.

Distributed File Systems ? Impressive. What about Sysplex ?

Cloud Computing ? 360/67 ; Virtualisation; Workload mgt; dynamic partitioning; ...

Well, try first the 10 volumes of ABC for zOS just for your curiosity.

Re:No issue here, Read the Patent!

[Anonymous Coward]

The real question should not be if it is dumb, but should the USPTO be giving a patent for something that probably has already been thought of, more specifically, is this an idea or method that will be created by other people? That is why a lot of these patents should not be awarded, they do reject patents (use to) by using common sense.

I like reading comments like yours as you are trying to calm the hysteria, and educate.

Re:No issue here, Read the Patent!

[Anonymous Coward]

Would you mind repeating that, but bolding the part which you think is actually innovative and patentable?

Re:No issue here, Read the Patent!

[gl4ss]

yeah so it's the thing but with "in googles cloud platform" in the end? that all the chunks are checked is kind of obvious. that the recycle bin gets deleted when the user is near the quota as well. deleting all the chunks of a file when deleting a file is just.. well.. eh.. like spesifying that when you eat the whole hamburger you eat the patty and both the buns.

for the user it sounds a bit of "we'll delete the files like, whenever!".

Re:No issue here, Read the Patent!

[Anonymous Coward]

If you actually read the patent, it is specifically for a similar method, but designed for Distributed File Systems.

aah... so it's a 'does something identical to an existing patent... but in the cloud' patent?

Re:No issue here, Read the Patent!

[sydneyfong]

No, but the fact that the owner of the Patent has a name that rhymes with poodle does.

I mean, seriously.

Re:No issue here, Read the Patent!

[Jafafa Hots]

The law is that it cannot be obvious to anyone "skilled in the art," not the layperson.

In other words, this cannot be obvious to programmers who work with files on distributed systems. It's doesn't mean that if it's not obvious to the average McDonald's worker, you're good.

Re:No issue here, Read the Patent!

[serviscope_minor]

Can we please have an end to the stupid articles where someone intentionally mis-interprets the abstract or even just the title of a patent and pretends it's some simple thing that's been done for decades to try to drum up anti-patent sentiment? There seems to be one a week or so.

Not until we have really stupid patents. I'm not a DFS guy, bu I am a computer guy. In every patent article there's one of you pointing out some supposed novelty. In my field, I've been through one or two and posted blow-by-blow rebuttals to slashdot pointing out that every claim is trivial or preexisting. Honestly, this patent has the same kind smell to it.

Deleting files based on last modification is not new. I think we can all agree on that and the cluster I used to run on 10 years ago had a script which did that. It wasn't new then.

They've basically munged up a time to live, and last modified time.

And patented the idea that "on a DFS you can delete it using some heuristic based on TTL, last modified time and quota".

All the blah about pluralities of chunks and pluralities of filesystems is patentese for obfuscating the underlying trivial point. The fact that it's a DFS is also beside the point, since there's nothing specific about a DFS there except for that obfuscaion.

It's a 1 line perl script "yeah but on a steam engine^W^Wa computer^W^Wthe internet^W^Wa phone^W^Wthe cloud^W^Wa DFS".

Re:No issue here, Read the Patent!

[serviscope_minor]

Well apparently I can't type.

Should read: not until we stop having really stupid patents.

And: I'm not a DFS guy, I'm a computer vision guy.

Re:No issue here, Read the Patent!

[Anonymous Coward]

Clearly no IBM training
Look at IBM's SMS - set a policy and its done. Set a policy to change policies. done.
Separate class for temp files. Done . Separate class for temp files that need extra security shredding - done .
Headers on files. done. Stacked files on one device done. Page stealing or recovering and recataloging deleted/expired files - done.
Dealing with mulit-volume files - done . Not enough? IBM has alter flags and last update, incrementl and base flags. All bases covered.
And if you thing this is good, wait till you see memory management and scheduling, where you have pre-canned whatever you want,
TTL, LSU and more. Add storage prediction reports. Tuning on user 'frustration' levels. done.When was this done? - circa 1960, and I believe MIT and other top American universities had all permutations covered. The Xephon manual on IBM shareguide will have
REXX scripts to delete files based on name in there too.
ON Mainframe , Midrange (ICL, Burroughs, Prime or PDP11/Vaxes this has been done. Did I mention attributes can be assigned to sms files - ie automatically compressed, mapped to HFS, or encrypted.

Re:No issue here, Read the Patent!

[Ginger Unicorn]

Goodle?

Re:No issue here, Read the Patent!

[Bill_the_Engineer]

Yes because none of us use similar file system naming scheme to handle our large amounts of data on our distributed file system. /sarcasm

Re:No issue here, Read the Patent!

[Theaetetus]

Can we please have an end to the stupid articles where someone intentionally mis-interprets the abstract or even just the title of a patent and pretends it's some simple thing that's been done for decades to try to drum up anti-patent sentiment? There seems to be one a week or so.

Unlikely. Nonetheless, anti-patent sentiment is a good thing. Far too many people assume there's some sort of fairness or justice to the whole mess, and there isn't.

But unfortunately, when the anti-patent crowd goes screaming to their legislators that we need patent reform because all software is obvious, or because Apple patented rounded corners, or because Google patented a 1970s mainframe, those legislators roll their eyes, nod politely, and call for security. There are some reasonable arguments for patent reform, but they get lost under all this FUD, discrediting anyone who tries to bring up those arguments.

Honestly, it makes you wonder if some of the anti-patent crowd on Slashdot is secretly pro-patent.

Re:No issue here, Read the Patent!

[Theaetetus]

If you actually read the patent, it is specifically for a similar method, but designed for Distributed File Systems.

Ahhhh... that's good.

You see, I was scared shitless that we are still quibbling over patents granted with the only claimed difference over some old methods (patented or not) being "on a computer". I see now how wrong I was: we stepped in the glorious era of the "in the cloud" claims.

More FUD. There are no patents that claim "[old method], on a computer". You can paraphrase some patents that way, as the Summary does where it says this is "essentially a kludgy variation." However, by definition, when you paraphrase something to explain it, you are explicitly replacing some of its terms with known, simple, easy to understand terms. For example, if I were to paraphrase a Tesla Model S to someone in 1950, I'd describe it as a sedan, but with electricity instead of gas. Does that mean that suddenly that's all a Tesla is? Could that 1950s person build one? Is a patent covering its power train suddenly trivial, because it's just "car, on electricity"? Of course not.

Paraphrasing is not the patent. The title is not the patent. Even the abstract is not the patent. The claim is the patent, and unless you can find a claim that says "1. A method, comprising doing a known method, but on a computer", then this is all just FUD and misdirection.

Re:No issue here, Read the Patent!

[Theaetetus]

Utility, novelty, and non-obviousness. This patent clearly fails at least one of these conditions.

"Guilty. The defendant clearly is guilty, because I say so."

You've presented a conclusion... how about some evidence to support it?

Re:No issue here, Read the Patent!

[c0lo]

If you actually read the patent, it is specifically for a similar method, but designed for Distributed File Systems.

Ahhhh... that's good.

You see, I was scared shitless that we are still quibbling over patents granted with the only claimed difference over some old methods (patented or not) being "on a computer". I see now how wrong I was: we stepped in the glorious era of the "in the cloud" claims.

More FUD

FUD? FUD you say? "You Keep Using That Word, I Don't Think It Means What You Think It Means"

I'm trying to scare nobody from nothing and neither cast any doubt on the patents on software (to me, it's clear: they are crap. You believe what you want about them).
I'd see acceptable for you to say about my post "That's shit... and of a bull origin!"; at least you'd be entitled to your own opinion and I'd try to see how you argue it. But FUD? That's so wrong, I stopped reading there.

Re:No issue here, Read the Patent!

[Theaetetus]

If you actually read the patent, it is specifically for a similar method, but designed for Distributed File Systems.

Ahhhh... that's good.

You see, I was scared shitless that we are still quibbling over patents granted with the only claimed difference over some old methods (patented or not) being "on a computer". I see now how wrong I was: we stepped in the glorious era of the "in the cloud" claims.

More FUD

FUD? FUD you say? "You Keep Using That Word, I Don't Think It Means What You Think It Means"

I'm trying to scare nobody from nothing and neither cast any doubt on the patents on software (to me, it's clear: they are crap. You believe what you want about them). I'd see acceptable for you to say about my post "That's shit... and of a bull origin!"; at least you'd be entitled to your own opinion and I'd try to see how you argue it. But FUD? That's so wrong, I stopped reading there.

Translation: "I have no actual response to your substantive comments, so I'll cut off your post at the second word and quibble about that one word. Hopefully, no one will notice."

Re:No issue here, Read the Patent!

[dajjhman]

After doing a quick reading up on Sysplex and the 360 systems, they seem to have managing shared temporary resources down but I can't find any reference to handling multiple chunks of distributed data per file being handled differently. Is there a more detailed resource I can find online? I am rather interested in looking deeper into it. Most reference I can find regarding their management of multiple systems is access to shared resources, not handling of individual distributed chunks of a file rather than the file as a whole. As for zOS it looks like a distributed file system, but again need more information to see how it handles the individual chunks of data rather than the files as a whole. The clarification is that the Google patent is for deletion of chunks of a file, such as a user profile, rather than the file as a whole.

Re:No issue here, Read the Patent!

[dajjhman]

The IBM systems have policies for managing shared resources and files yes, but for an entire file or resource. The Google patent is specifically for handling the individual chunks of a distributed file. ex) a user has their profile stored in a file, let's call it user.xml. Each individual user.xml file is comprised of chunks (like , , , etc). let's say that the portion of the file gets updated more often than the portion. Rather than setting a policy for the entire file to have a shorter TTL in cache, you set only the TTL of the block to be very short while having a longer TTL for the chunk. I've yet to find a reference to an IBM system doing this on such a low level. If anyone can share a link to prove otherwise please do.

Re:No issue here, Read the Patent!

[c0lo]

Translation: "I have no actual response to your substantive comments, so I'll cut off your post at the second word and quibble about that one word. Hopefully, no one will notice."

Would your second comment come with a rectification of the misuse of FUD (I don't know... probably some apologies for misusing it) maybe I'd be inclined to consider what comes after the offending word.
With the lack of even an attempt for rectification and even adding a bit more... insomnia or not, I'm not going to waste the time to read some opinions pushed a second time with a pretense of significance. I hope you'll find the satisfaction of others noticing your comment and give it more attention than I did, maybe it deserves it.

Have a nice day ahead mate

Re:No issue here, Read the Patent!

[Theaetetus]

Translation: "I have no actual response to your substantive comments, so I'll cut off your post at the second word and quibble about that one word. Hopefully, no one will notice."

Would your second comment come with a rectification of the misuse of FUD (I don't know... probably some apologies for misusing it) maybe I'd be inclined to consider what comes after the offending word.

Clearly, you're the real victim here. Forget your threadjacking and obfuscation of our topic, your refusal to address the substantive points in the post- no, you've been offended!

Well, my utmost apologies to you, Sir. I feel shamed for using a three letter acronym in a way that you felt was offensive. Your pearl clutching, swooning collapse due to those three letters is not unnoticed, and I will try to remedy my behavior in the future, should we ever meet again. I hope your blood pressure eventually returns to normal, and my condolences to your ulcer.

Re:No issue here, Read the Patent!

[sydneyfong]

Yes, as you can see, they're the Good people. :)

I know what you mean, probably could have phrased it better. Anyways.

Re:No issue here, Read the Patent!

[c0lo]

Clearly, you won the argument, I'm speechless in the face of your eloquence. My apologies for expressing my opinion, hijacking and obfuscating the otherwise a so interesting and fertile soil for insightful exploration on the matter of patents; I didn't realize a titan of patent topics is present on the thread to set the things straight, mea culpa.

I just hope I made your day feel worth living with my above admission, it seems winning is more important to you that carrying a dialog. Keep up the good work and, as a humble suggestion, spend less time on /. - it may interfere too much with your patent legal career.

(btw: I argue the patent is obvious and "a plurality of previous art exists". You only need to look for "rolling logs" and you'll see the same principle applied there; the fact that one is distributing the logs - or any "file segments" - over multiple devices/nodes/filesystems/tapes/whatever and still manages them without the need of an external "library index" or "database", but only following conventions on location/naming - is exactly what I accused in "obvious patents with the extra in the cloud claim")

Re:No issue here, Read the Patent!

[Anonymous Coward]

i dont see what is wrong with patenting software that performs a specific specialized task involving something like a file format that is not open source, and CLEARLY proprietary involving many many many hours of eye bleeding work to take from conception to implementation.

but patenting a subroutine that performs a generic task like deleting temporary files? oh but they files are split up into chunks whoopdy fuckin do, remember the assholes that try to patent shit like a linked list?? they should be exiled to leper colonies. in my mind this type of bogus crap equates to a golfer patenting his swing or some dumb shit like that. protect the actual innovators not big corporations who would like to patent generic every day type shit. why would anybody defend this kind of bullshit unless they don't actually understand how this could easily get out of hand over the next 40-60 years and programming without corporate armor backing you will be impossible unless you're in it just for fun.

and yeah there's more than one way to skin a cat, but when it becomes impossible to skin a cat without having googles permission what the hell is going to happen to cat hat prices?? how much control would you have if you owned all the patents that the 'futurenet' is built upon.

Re:No issue here, Read the Patent!

[nnnnnnn]

Software should not be patentable. Period.

Right, software that runs the entire modern economy should have 0 IP rights given to those who create it, meanwhile swinging on swing gets 20 year IP protection.
http://www.google.com/patents/US6368227

Re:No issue here, Read the Patent!

[chrismcb]

This really has nothing to do with DFS. The ONLY difference between this and a regular FS, is in a regular FS you look at one file and get the timestamp. in the DFS you look at all them, select the latest.
It isn't about it being done before (although I'd be shocked if it hasn't) Its about how novel it is. This patent is basically saying, look there is 5% of the diskspace left... This file is too old, delete it. Ohh and on a DFS. If it is brown, on a thursday, and frogs are raining.

Re:No issue here, Read the Patent!

[chrismcb]

It's a patent on storing files.

Well, actually it is a patent on deleting files.

Those who don't learn from history

[jcrb]

are doomed to think they have (re)invented it.

This is so true I have quite a few patents and I see it every day while doing art searches the number of patents claiming things that anyone with even a half way decent understanding or education in the field would recognize as already having been done "way back in the good old days".

Next, the Generation Data Set!

[Gim Tom]

Wow, Has anyone patented the concept of the old mainframe Generation Data Set recently? I used them extensively back in the mainframe days and could have used a similar concept in more recent systems, but never found a real substitute in either Unix/Linux or Windows. A simple explanation for those who have not heard of them is that they are sort of a push down stack of files managed by the OS with a fixed stack length. You could reference them by a long serial number that was of the format GnnnnVnnnn or by a simple index where the top (and most recent) file in the stack had a zero index. Has anyone seen anything similar in either Unix or Windows?

Re:Next, the Generation Data Set!

[Anonymous Coward]

OpenVMS has its versioning file system. It's similar. Full file names are expressed as name.extension;version

Re:Next, the Generation Data Set!

[FrankSchwab]

Ahh, those halcyon days of youth on VAX/VMS turning in Assignment4.c;97

Those were the good old days.

Re:Next, the Generation Data Set!

[fatphil]

There was a C compiler? All I remember on the system was emacs, scheme, and mail (probably news too).

What's old is new again

[ZorinLynx]

The same thing happened in the 80s and early 90s when microcomputers started gaining features like virtual memory, protected modes, out of order execution, etc... People thought these were all brand new things, when in fact mainframe processors had done all that 20 years prior in the 1960s.

I bet when all the kids were super-excited about programming on the i386 with its "OMG VIRTUAL MEMORY!!!" the older guys who had worked on mainframes just rolled their eyes. :)

Read about the IBM 360/91 if you want details on what I mean. It was amazing when you consider the year it came out.

Re:What's old is new again

[Anonymous Coward]

The same thing just happened in the last few years when hypervisors and virtual machines were "discovered". IIRC IBM invented the hypervisor/vm concept some time in the late 50s and commercialized it in the early 60s.

You never see the bare metal on a mainframe or even a pSeries UNIX server. You are always running under the hypervisor.

Re:What's old is new again

[GoChickenFat]

I bet when all the kids were super-excited about programming on the i386 with its "OMG VIRTUAL MEMORY!!!" the older guys who had worked on mainframes just rolled their eyes. :)

You talking about the same old grey beards that gasped when the kids opened the cover on a server and added their own memory, network adaptors, backplanes, disk drives, etc without having to call IBM out to do it?

Re:What's old is new again

[Anonymous Coward]

IBM 360, oh the memories! *shudders* Let's see, sleeping bag *check*, back pack *check*, change of clothes *check*, food for a couple days *check*, some beef jerky and crackers just in case *check*, thermos of hot coffee *check*, textbooks, notebooks and loose paper *check*, flow charts and punched cards *check*, ok, time to go get in line at the student card reader.

Re:What's old is new again

[sbjornda]

I bet when all the kids were super-excited about programming on the i386 with its "OMG VIRTUAL MEMORY!!!" the older guys who had worked on mainframes just rolled their eyes. :)

You talking about the same old grey beards that gasped when the kids opened the cover on a server and added their own memory, network adaptors, backplanes, disk drives, etc without having to call IBM out to do it?

Yeah, and then rolled their eyes again because the kids didn't know about change control, didn't notify the users about the outage, didn't verify that their backups were good (if they even had backups), and lost 6 months worth of corporate data as a result.

--
.nosig

Re:What's old is new again

[Daniel Dvorkin]

I bet when all the kids were super-excited about programming on the i386 with its "OMG VIRTUAL MEMORY!!!" the older guys who had worked on mainframes just rolled their eyes. :)

Well, it was super-exciting to have it on the desktop for a reasonable price, yeah. I can't speak for everyone of that generation, but I appreciated it while still understanding perfectly well that it wasn't a new invention.

Re:What's old is new again

[93 Escort Wagon]

As an aside... I remember a few years ago - when we were still running tape backups - I went to one of our then-sysadmins and asked him to recover an important directory one of our faculty had managed to delete. I was told he couldn't do it because it would require they stop the backup system for several hours, which would throw their backup tape rotation scheme out of sync.

So we were continuously generating backups we could never actually use.

Re:What's old is new again

[stox]

Close, invented in the mid 1960's, CP67 on the 360/67, commercialized as VM370 on the 370 in 1970.

Re:What's old is new again

[FrankSchwab]

In all my years of corporate computing, I've gone to IT 4 times to get a file off backup (4 different IT departments at 4 different companies). My success rate? 25%.

Each time, the answer came back "I'm sorry, but we weren't able to recover that file".

For some reason, I have a distrust for IT Backup, and make my own copies of important stuff. /frank

Re:What's old is new again

[dbIII]

You'll need to clarify this - never actually use or just not deliver instant gratification?
Also would instant gratification be available if you had escalated it up the tree to somebody that could take responsibility for them having to throw their backup tape rotation scheme out of sync?

There's a big difference between a fucked system and just not being prepared to drop everything at the whim of anyone. Which was it?

Don't go attacking me on this, I've got stuff on disk as well as tape to provide that instant gratification to anybody that asks, but in other places there isn't always that luxury so I see how there can be delays.

Re:What's old is new again

[DigiShaman]

Not a backup, but shadow copies are an admin's friend. I've had to refer to those 99% of the time when a file gets deleted or corrupted by an end user. Otherwise, It's back to restoring from a grandfather-father-son backup.

Not sure about in Linux, but in the Windows world, backup software just plain sucks balls! The built-in backup utility for Server 2008 is already functionally obsolete (wont support larger sector sizes on external or GPT volumes as a destination). Symantec Backup Exec is always a hair-on-fire experience. Microsoft DPM will backup, good luck restoring data. Acronis is a half-baked solution with a pretty interface...etc.

Re:What's old is new again

[Anonymous Coward]

No, not those old grey beards. The ones he was talking about knew how to do that. Many of them soldered together their own computers from hobby magazines (young whippersnappers don't know what a PCB is, let alone a logic gate).

Sometimes they worked on company servers where the SLA required that the vendor perform the upgrades. Big deal.

Re:What's old is new again

[93 Escort Wagon]

No, seriously - he didn't want to give me the files *at all*. I even told him I could wait a couple days... but it really was a bizarre case where he felt these tape backups (which, at the time, were our ONLY backups) were only for use if the building fell down. In that case, the tape set that had been rotated off site would be brought in (after purchasing all new tape drive and server hardware, of course), and used to reconstruct our servers.

Fortunately nowadays we have actual competent (and pragmatic) sysadmins, and we also have disk backups. And if I need something, I can retrieve it myself.

Re:What's old is new again

[Anonymous Coward]

Pffft... It was only a few 1's and 0's.... can't some one just type them back in?

Re:What's old is new again

[Anonymous Coward]

That's funny - our mainframe crew had been backing up to tape for years but failed to put the catalog files on tape so they never could be restored. In every disaster recovery test they had to pull the catalog from the "down" mainframe in order to restore the tapes at the DR site. ...and never once did they think that was a problem.

Re:What's old is new again

[rjr162]

We use Tivoli Storage Manager (TSM) for all of our backups

Re:What's old is new again

[dbIII]

So what did management say about it? Or did you just leave it at that?
Sorry, it just seems really unlikely unless the person did not think it was their job to give you the files as set by some kind of possibly stupid policy that could only be overruled by somebody high enough in the org to set the policy. I've seen that sort of shit but it's rare outside of places that are either dying or too big to fail zombie orgs.

Re:What's old is new again

[LunaticTippy]

I used Backup Exec for many years and never had any complaints about it. It was tricky to get the tape and library drivers right (don't use the manufacturer's ones, use Backup Exec's) but once I got everything humming it worked great. I set up daily, weekly, monthly, and manual backups and managed tape rotations. I kept a set of onsite disk backups so I could restore files deleted in error quickly, and the software would automatically overwrite the oldest ones with no oversight. I spent very little time administering it, and restored countless files as well as one entire server no sweat.

Now we're forced to use Networker. I'm not impressed so far, stuff that was intuitive before now requires me to look up how to do it. I've also run into a bug where it appears to be running a restore but in reality will never complete. No error message.

Re:What's old is new again

[Anonymous Coward]

Had a few nice stories of customers with backup solutions. One was the customer who discovered that he could just press Return when the backup program prompted him to insert the next cartridge of the series.

Another was the one who backuped daily for years on the same cartridge, never changing it. When there was an actual need to restore from backup, the tapes were blank. Nothing magnetizable left on them.

Re:What's old is new again

[Anonymous Coward]

You mean like all the "hype" (and I wonder how many patents they applied for?) over "virtualization" (VMWare, etc)... like hey! You can spin up a bunch of virtual "machines" on one larger machine!! (cough cough, IBM VM/370... we were doing that on a 3033 with 16M of physical memory in the early 80's).

But, see, now we can have multiple machines doing that (instead of one that filled an entire room, with water cooling...), and call it a "cloud"!!

No they didn't

[Anonymous Coward]

No they didn't. Below is what they patented. In order for your statement to be true - EVERY single bit of the description below would have to be included in that 70's mainframe you are talking about. I have a hard time believing that back you the 70's you were dealing with files that where divided into a plurality of chunks that are distributed among a plurality of servers.

"A computer-implemented method comprising: selecting a file having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers, wherein each chunk has a modification time indicating when the chunk was last modified, and wherein at least two of the modification times are different; identifying a user profile associated with the file; determining a memory space storage quota usage for the user profile; deriving a file time to live for the file from the path name; determining a weighted file time to live for the file by reducing the file time to live by an offset, where the offset is determined by multiplying the file time to live by a percentage of memory space storage quota used by the user profile; selecting a latest modification time from the modification times of the plurality of chunks; determining that an elapsed time based on the latest modification time is equal to or exceeds the weighted file time to live; and deleting all of the chunks of the file responsive to the determining. "

Re:No they didn't

[Khyber]

Ahh, yep. UNIX had that and more long ago, as did WYSE system shit.

BS patent

Re:No they didn't

[Anonymous Coward]

No they didn't. Below is what they patented. In order for your statement to be true - EVERY single bit of the description below would have to be included in that 70's mainframe you are talking about. I have a hard time believing that back you the 70's you were dealing with files that where divided into a plurality of chunks that are distributed among a plurality of servers.

Perhaps not. But that's COMPLETELY IRRELEVANT. The technique is no different on a distributed file system versus a non-distributed file system. Given that this technique was used on a non-distributed file system, its application to a distributed file system is obvious. In other words if the only thing new is the application of an old technique to a new file system, it's an "exhausted combination claim".

Re:No they didn't

[Qzukk]

In order for your statement to be true - EVERY single bit of the description below would have to be included in that 70's mainframe you are talking about

In order for your statement to be true, the Doctrine of Equivalents would have to be eliminated.

Oh bullocks

[the eric conspiracy]

The summary is wrong. Folks, please stop reading the abstract, and read claim 1 instead.

This is what is patented:

1. A computer-implemented method comprising: selecting a file having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers, wherein each chunk has a modification time indicating when the chunk was last modified, and wherein at least two of the modification times are different; identifying a user profile associated with the file; determining a memory space storage quota usage for the user profile; deriving a file time to live for the file from the path name; determining a weighted file time to live for the file by reducing the file time to live by an offset, where the offset is determined by multiplying the file time to live by a percentage of memory space storage quota used by the user profile; selecting a latest modification time from the modification times of the plurality of chunks; determining that an elapsed time based on the latest modification time is equal to or exceeds the weighted file time to live; and deleting all of the chunks of the file responsive to the determining.

Re:Oh bullocks

[Anonymous Coward]

Patent hint: Needs to be not obvious to one skilled in the trade...

Re:Oh bullocks

[Anonymous Coward]

This incomprehensible POS filing blows a "plurality of chunks that are distributed among a plurality of servers". Who can possibly make sense of this ridiculous prattle?

Re:Oh bullocks

[Anonymous Coward]

What they've patented here is actually the per-user tiering or enforcement of storage quotas by deletion, weighted by the personal usage, using time as a preference/priority measure. It's date-based expiration, but scaled based on disk pressure. If I've got a lot of data, it means that data with the nearest expiration date *scaled by some function of the memory pressure*, is deleted.

The summary makes it sound more trivial than it is, since it takes a short snipped from the example usages section of the documents....

Re:Oh bullocks

[edibobb]

...and unique.

Re:Oh bullocks

[Anonymous Coward]

My dick responded to your determining.

Re:Oh bullocks

[bill_mcgonigle]

If the patent process were anything like the peer review process, a bunch of distributed filesystem engineers would have been asked how to implement file expiration, and their answer, within five minutes, would sound something very close to this.

But, more importantly, Google seems to have actually implemented this (not bad, considering the state of things). But who honestly believes they would not have done so without the hope of patent protection?

Re:Oh bullocks

[bill_mcgonigle]

If I've got a lot of data, it means that data with the nearest expiration date *scaled by some function of the memory pressure*, is deleted.

So like any of the thousands of scripts sysadmins have written to check the output of df and run an rm on the temp directories with a -mtime value to find based on the result of the df. Hint: they're quite useful for maintaining large cache trees.

To anticipate the next patent: only take the action immediately if the return value from df is on the high side, otherwise schedule the pruning activity to minimize impact on CPU and IO resources (I shall call my inventions, 'nice' and 'ionice'). Throw some 'pluralities' in there too, for good measure.

Re:Oh bullocks

[Anonymous Coward]

Patent lawyers. The patent system is run by patent lawyers for patent lawyers (and managers). Engineers are not relevant to them, despite the high justifications of the patent system supposedly enabling innovations to be published for the edification of other inventors.

Re:Oh bullocks

[prefec2]

We were thought similar features in university. To be precise I attended courses on operating systems, which included file systems. File systems can be distributed and files can be distributed. Especially in file systems which virtualize the block layer. The only new thing is that they do not use the newest modification time as last modification time, but established a live time value. So the new thing is "weighted file time to live". The rest is again the method to determine the most outdated file. In the end they are deleting chunks of a file, which would normally corrupt a file beyond recognition, if it is a binary file. Therefore, there type of file is either a old mainframe type of file, based on records (which were the predecessors of databases) or if chunks are not records or sequences of records, then it is just a bunch of files bundled together, where single files can be deleted.

If a chunk (which is not really a precise term) is an arbitrary long segment of bytes in a file, then any deletion must honor the file structure in a way, that the inner structure is not corrupted. This is possible when you have well defined boundaries for your chunks. Which is then only something similar that (for example) git does with file deltas with the addition that only sound chunks are accepted.

Re:Oh bullocks

[Anonymous Coward]

The summarry is fine. The excess verbage in the patent should not confuse anyone. I suspect there is prior art as well since the concepts of having files distributed on many devices and automated deletion of expired files have existed for some time now.

Re:Oh bullocks

[serviscope_minor]

What I hear from you is blah blah blah.

Don't you get it? This is IN THE CLOUD!

So it must be new.

Re:Oh bullocks

[fatphil]

"identifying a user profile associated with the file"

They'll never get any royalty money from me, then. My system is completely different. Mine identifies a group profile associated with the file. Yes, I only have one user in each group currently, why do you ask?

Re:Oh bullocks

[Theaetetus]

If the patent process were anything like the peer review process, a bunch of distributed filesystem engineers would have been asked how to implement file expiration, and their answer, within five minutes, would sound something very close to this.

If the patent process were anything like the peer review process, patents would name dozens of "inventors" based on their seniority and tenure, and would be granted for political reasons rather than their novelty.

Re:Oh bullocks

[Anonymous Coward]

The patent process *isn't* like the peer review process. The criterion for obviousness is from the perspective of one of *ordinary* skill in the art. "A bunch of distributed filesystem engineers" are of far greater skill than that. All sorts of stuff might seem obvious to the filesystem engineers, but still wouldn't be obvious in the eyes of the law.

Re:Oh bullocks

[chrismcb]

The summary is wrong. Folks, please stop reading the abstract, and read claim 1 instead.

Why? You do realize pretty much everything in claim 1 is in the abstract, and worded better:
The abstract:

• The file is associated with a time to live derived from a path name for the file.
• The file is divided into a plurality of chunks that are distributed among a plurality of servers.
• Each chunk has a respective modification time indicating when the chunk was last modified.
• A latest respective modification time among the respective modification times of each of the plurality of chunks is selected.
• A determination is made as to whether an elapsed time based on the latest modification time is equal to or exceeds the time to live.
• Each of the chunks of the file is deleted responsive to the determination.

While the first item is similar to the IBM patent, the rest of the items are... well just not very novel.

US Patent and Trademark Office Dumb Asses

[CuteSteveJobs]

If you screwed up like this in company, you would be fired. Yet some dumbass government worker in the USPTO grants this and several million dollars later it gets sorted out by the courts. One of the reason patent litigation is out of control is because these dumbass don't do their jobs.

Worse, the USPTO is about to switch from first to invent to first to file. You don't need to invent any more. Just find out what your competitors are doing, patent it, and sue them out of business: http://www.jdsupra.com/legalnews/top-ten-reasons-to-file-your-patent-appl-98912/

Re:US Patent and Trademark Office Dumb Asses

[Anonymous Coward]

Wrong. The USPTO is switching to a first-INVENTOR-to-file system. You can't have derived the invention from your competitors.

Re:US Patent and Trademark Office Dumb Asses

[stevesliva]

and several million dollars later it gets sorted out by the courts

No, not really. Several million dollars later, someone capriciously wins, and then there are equally capricious rounds of appeals.

Re:US Patent and Trademark Office Dumb Asses

[Daniel Dvorkin]

If you screwed up like this in company, you would be fired boilerplate anti-government yammering blah blah blah

When you get out of high school and have your first job that doesn't involve cleaning a grease trap, you'll learn how things actually work in the corporate world.

Re:US Patent and Trademark Office Dumb Asses

[Zontar The Mindless]

You forgot to put work in quotes. ;)

Re:US Patent and Trademark Office Dumb Asses

[Daniel Dvorkin]

Hah! Good point. :)

Re:US Patent and Trademark Office Dumb Asses

[dbIII]

The patent worker did not screw up since the paperwork was filed and the fee was collected. Validity has been shifted to the courtroom now.
So to sum up - the USPTO doesn't give a shit so has no reason to fire.

Public Patent Challenge

[cowtamer]

I think it's time for a crowdsourced patent challenge web site run by the USPTO where there would be a period of public comment for each patent about to be awarded in order to help underpaid (and I imagine under-resourced) examiners find Prior Art.

A lot fewer patents might be awarded, but ones that are would be genuinely new -- this might also save the world billions of dollars.

Re:Public Patent Challenge

[stevesliva]

I think it's time for a crowdsourced patent challenge web site run by the USPTO where there would be a period of public comment for each patent about to be awarded in order to help underpaid (and I imagine under-resourced) examiners find Prior Art.

A lot fewer patents might be awarded, but ones that are would be genuinely new -- this might also save the world billions of dollars.

http://peertopatent.org/

Re:Public Patent Challenge

[Anonymous Coward]

Don't forget to support multiple languages as there are a lot of countries that can constantly demostrate prior art to the US.

Computers - Poland/England
Rocketry - China/Germany
Car engine - Germany

Pretty much anything claimed by Edisson by produced by people like Tesla.... etc!

Re:Public Patent Challenge

[martin-boundary]

Along those lines, it would be a lot easier to simply limit absolutely the number of patents awarded each year, say to 100 across all disciplines and industries altogether. That would force a ruthless culling on merits.

Ideas are a dime a dozen. The patent office should collect fees from thousands of hopeful inventors, and spend a year deliberating behind closed doors, then announce the 100 patent winners at a ceremony, like the Nobel prizes (in science, not politics). Most inventors would lose out, but the benefit to society would largely outweigh the fees they paid.

Ask yourself why Nobel prizes (in science) are impressive and why they wouldn't be if there were thousands of them each year in every field.

Re:Public Patent Challenge

[Legion303]

I'm a lot more in favor of this scheme: any company found abusing the patent office with shit like this more than x times should have every one of their patents suspended for however many years it takes the backlogged USPTO to review them all thoroughly. In queue behind all pending USPTO operations.

Virtual memory, etc. was easy in the early 1980s

[davidwr]

Designing computer hardware that could do virtual memory and other features that were OMGNEW in the early 1980s was very easy. As you said, it had been done for decades.

Designing a CPU that would go in sub-$2000 computers that could do these cool things, on the other hand, not so easy.

ven if they are the first to do it.

[Anonymous Coward]

How many month of research investment is this patent supposed to be protecting, aren't patents supposed to for no obvious things. Sticking date info into a files name is five seconds of semi inspired thought, it's something that nobody would every publish in a software engineering book (except as a side note perhaps) because it's the sort of thing you would just reinvent as needed.

More than the sum of its parts

[Anonymous Coward]

And the inventive step is....
And the non obvious part is....

The problem here isn't the USPTO, it's the Patent Appeals Court that modified the Supreme Court decision (that an invention needed to be more than the sum of its parts), and decided that as soon as you'd been told about an invention, your judgement would be tainted by 'hindsight bias' and thus unable to determine prior art. So unless it's written down in that form, the patent should be awarded.

Can I ask the idiots in the Patent Appeals Court, is THIS invention more than the sum of its parts?

No?!

Re:More than the sum of its parts

[Grond]

What are you talking about? There is no "Patent Appeals Court" in the United States. There is the Court of Appeals for the Federal Circuit. And the standard for non-obviousness was most recently articulated by the Supreme Court in KSR v. Teleflex , a 2007 case in which the Court held that the precise prior art combination did not need to be explicitly "written down in that form":

As our precedents make clear, however, the analysis need not seek out precise teachings directed to the specific subject matter of the challenged claim, for a court can take account of the inferences and creative steps that a person of ordinary skill in the art would employ.

The Court further loosened concerns over hindsight bias:

A factfinder should be aware, of course, of the distortion caused by hindsight bias and must be cautious of arguments reliant upon ex post reasoning. ... Rigid preventative rules that deny factfinders recourse to common sense, however, are neither necessary under our case law nor consistent with it.

As for the "more than the sum of its parts" standard: that comes from the Great Atlantic & Pacific Tea Co. case of 1950. And it wasn't modified by some "Patent Appeals Court." It was statutorily overruled by Congress by the Patent Act of 1952, as later interpreted by the Supreme Court itself in Graham v. John Deere .

Re:More than the sum of its parts

[chrismcb]

THIS invention more than the sum of its parts?

No?!

Sure it is. IT doesn't help if you have a TTL in the filename, if you don't do anything with it. It doesn't help if the user has a file quota, that you let fill up with old dead files.
BUT that doesn't mean this is patentable. Its basically "delete the largest oldest file, when the user starts to run out of disk space, on a DFS" I have no "hindsight bias" but I have to believe something like this has been done before. But shoot it is pretty much how I would implement it, if asked... And that is the point of the "novelness" of the question.

Anyone else remember August 24, 1979?

[Anonymous Coward]

I Stapled your mom's ass that night.

Google do no evil

[Anonymous Coward]

Google can do no wrong or evil (bs?!?!?!)

Storing data

[destruk]

I'll file a patent for the storing and retrieval of data in some form. This would cover books, video, computers, and the human brain. Now, give me my payments.... hey,nobody has ever heard of this because there are no records before I filed this patent right?

I've recently been granted a patent...

[KNicolson]

...for extracting random phrases out of the middle of a patent document that match prior art and posting them to a web site in order to increase hit rates. Please delete this article or you will be hearing from my lawyers!

This is not-obvious? There is no prior art?

[Anonymous Coward]

This has to be the worst excuse for a patent ever heard. Automatic deletion of TMP files?

Next Google wont want to pay taxes!

(oh... snap...)

Except the summary is wrong

[tuppe666]

I make no claims to the validity of the the data, but the example given and the patent are different. The IBM 360 example is about *preserving files* by affording them additional protection, as opposed to the Google patent which is about *deleting* temporary files through adding a "time to live" value actiuallin in the directory/filename with various ways of cleaning out these files, as well as an *additional* indicator that it is a tempory file.

...but considering the patent is 26 pages long, if could be for walrus toothbrushes....and who the hell quotes from a book, and then posts a link to the whole book!

On the Internet!

[Anonymous Coward]

It's probably not google's fault. I wish patent applications cost more money, the more patents you had.
But big companies would find a loophole.
sigh.

Re:Virtual memory, etc. was easy in the early 1980

[jedidiah]

Miniaturization took care of that.

Whether or not it could be done for some arbitrary price is not relevant.

Doing something over again in a different medium is still not invention no matter how much you want to shill for companies that would grind you into crackers if given enough motive.

Except Google are not doing that.

[tuppe666]

I wonder if google is going to chase me. I use that exact method for log file expiration in a program I wrote back in 1998 for scanning configurations across servers, From memory I also got the code for doing that from someone elses web site that had posted the sample. the log files are written to dated folders according to how long I needed to keep a record for of the specific scan being executed.

I doubt it, as the use case is completely different, you *manually selected what to delete" on log files on criteria you created, As opposed to Google who indicate a file is *temporary* by appending a file suffix/ changing a bit/ Mimetype, and then deleting it when...

Re:Except Google are not doing that.

[sycodon]

It's a stupid patent and the examiner was an idiot.

Why are you defending these clowns?

Re:Except Google are not doing that.

[Patman64]

Their name starts with a G instead of an A or an M.

Re:Except Google are not doing that.

[rtfa-troll]

Mostly because, by putting up a false accusation in the original article, the poster became vulnerable to this type of comment. Remember there are a number of people who benefit from patents by ensuring that it's difficult for competition to enter the software market. There are plenty more who believe they benefit even when they are actually getting lower paid jobs because of this. If someone makes a stupid accusation these people will always step in to show that it was wrong.

GROW UP

[Anonymous Coward]

geez, when is slashdot ever gonna stop running these stupid articles that only show how little the posters know about patent law
or, at least, READ THE FILE WRAPPER 111
MAYBE THE IBM PATENT IS AN X OR Y DOC IN THE SEARCH REPORT 111
OR THE VERY LEAST, READ THE CLAIMS !!!

claim 1:
  A computer-implemented method comprising: selecting a file having a path name in a distributed file system, wherein the file is divided into a plurality of chunks that are distributed among a plurality of servers, wherein each chunk has a modification time indicating when the chunk was last modified, and wherein at least two of the modification times are different; identifying a user profile associated with the file; determining a memory space storage quota usage for the user profile; deriving a file time to live for the file from the path name; determining a weighted file time to live for the file by reducing the file time to live by an offset, where the offset is determined by multiplying the file time to live by a percentage of memory space storage quota used by the user profile; selecting a latest modification time from the modification times of the plurality of chunks; determining that an elapsed time based on the latest modification time is equal to or exceeds the weighted file time to live; and deleting all of the chunks of the file responsive to the determining

to be covered by claim one, you must meet each AND every part..
did IBM, back in 1970s , describe a file divided among two or more (= plurality) servers, where at least two of the parts has a diff modification time, and dong the storage quote athing ??

patents are specific; i leave it to the experts to say if this one is worth anything

NCAR Mass Store had expiration dates

[cruff]

The NCAR Mass Store (tape archive) had an expiration period attribute (units of days) on the bitfiles. The default, if not specified was 30 days, which effectively made it a temporary file. Expiration periods of 31 days or more were considered more permanent, and the owners would receive email two weeks and one week before the projected expiration date arrived. Expiration processing was run each Sunday, and the bitfiles were moved into the trash, from which they could be recovered for another 30 days before they were permanently deleted. This was in the mid-eighties.

shouldnt be modded down

[Anonymous Coward]

Regardless of his intent, The truth is there are many americans who have spanish as their first language.

USPTO's new slogan

[Stirling Newberry]

"We don't even pretend to care."

Google is trying to patent

[Stirling Newberry]

A one liner find command that has been written thousands of times, if not millions of times.

Not novel, not original, prior art, obvious.

read the eff'n claims, not the description

[Anonymous Coward]

The claims on this are fairly narrowly drawn..
Everyone writes the description/disclosure broadly (check out the number of patents titled something like "catalytic improvement") to establish "prior art" for folks following, and then you write the claims both broad and narrow. broad claims get knocked out by the examiner, but you keep the narrow ones, so nobody can 'duplicate' your work.

There is a patent on a banana peeling machine which describes all sorts of ways one can peel bananas, but the claims cover only one specific implementation with a specific number of blades, etc. That's enough to
a) prevent someone from making an exact knockoff of YOUR banana peeling machine
b) patenting some other banana peeling machine that you might invent in the future.

Been doing this for years...

[bl968]

find /tmp/* -mtime +14 -exec rm {} \;

Re:Been doing this for years...

[jcdr]

Mod parent up.

No need to add a new date information in the file name. The inode already have the creation, last modification and last reading date. Fare enough to determine witch files are too old.

Re:Been doing this for years...

[Anonymous Coward]

It's not enough, because the file can't tell you, individually, whether it is too old.

Trivially, this example says 'delete all files that are X amount old'. Not 'delete all files which say they have expired'.

Re:Been doing this for years...

[jcdr]

The inode information are individual to each file.

Either the purpose is to let system width cleanup regardless of the application that created the files, in witch case inode info is the only practicale soultion because there not grant that every applications will uses the same convention.

Either the purpose is application specific in witch case the responsibility to remove the file belong to the application, not to the system tools.

Re:Been doing this for years...

[chrismcb]

But you aren't taking into consideration the amount of disk space the user has left. You are just unilaterally deleting everything after a certain date. No patent for you.

It's not old til you've married it.

[Ostracus]

Hey, things are new if you've never seen them before!

Just wait till you all grow up and discover cougars.

Re:It's not old til you've married it.

[n6kuy]

When I'm all growed up, it'll be too late to discover cougars.

Re:Virtual memory, etc. was easy in the early 1980

[Anonymous Coward]

Designing a CPU that would go in sub-$2000 computers that could do these cool things, on the other hand, not so easy.

Ring us back when you can distinguish between implementation and innovation.

I'm Sorry, but...

[sycodon]

But the USPTO is populated by idiots.

They are deserving of the disdain and ridicule reserved for the Postal Office, Congress, etc.

Which is a shame because I've always figured they had some pretty smart people there. The examiner should have taken a shit on the application and mailed it back with a note saying,"this is what your application is worth".

They are either complete morons or...are getting payoffs. And Google will just use it as club some day on a small outfit that doesn't have half a million dollars to fight a lawsuit.

Re:I'm Sorry, but...

[jatoo]

They are either complete morons or...are getting payoffs.

Or they are precisely following moronic policy

Re:I'm Sorry, but...

[jockm]

Or they are normal people, without much domain knowledge, forced to handle too many cases in too little time, and fit within the rules of a broken system.

I personally find that to be the more plausible situation.

Re:I'm Sorry, but...

[Lotana]

Or maybe they are not experts at the field of computing. Certainly not much history about our field is being recorded, especially in easily accessible form.

Imagine if someone give you to approve a patent application on some specific aspect of rocket design. How would you approach to researching this, when you got thousands more applications to process with tight deadlines?

Plus we only hear about when obvious troll submissions get through. I certainly don't know the ratio of good and bad patent grants. Alas as far as patents are concerned, Slashdot crown will accept nothing but impossible perfection.

Re:I'm Sorry, but...

[Anonymous Coward]

I'm not even from the US but I'm sure the USPTO has thought of this since the start. They had to assume that the patents will come from all fields of technology so I don't expect there to be 20 people, the same 20 people, discussing ALL patents. I'm sure there are committees per various fields of science and technology.
I'll even wiki it AFTER I have posted this, that's how sure I am of this :P

Re:I'm Sorry, but...

[reve_etrange]

The USPTO is supposed to support itself with fees. The largest fee is for reexamination, creating a financial incentive to grant bad patents (which are likely to be reexamined). -da

Re: I'm Sorry, but...

[Anonymous Coward]

You mean you will alter the wiki to fit your needs post-posting not pre-posting? I salute you Sir.

Re:I'm Sorry, but...

[Anonymous Coward]

No there are no committees, at least not making decisions on whether to grant specific patents. Eac patent application is handled by precisely one patent examiner who gets something like 20 to 40 hours to handle the entire process from finding prior art to writing a decision and addressing the appeals the applicant is likely to make.

Re:I'm Sorry, but...

[AmiMoJo]

Google only uses patents defensively, at least up until now. In a way it is better that such a ridiculous patent went to a non-troll company that won't use it to suppress the competition, if the USPTO is going to grant such nonsense.

Re:I'm Sorry, but...

[ChrisSlicks]

When it comes to software or computing patents they apparently spend 90% of that allotted time by playing mine sweeper.

Re:I'm Sorry, but...

[tommeke100]

Sure, but automatically deleting temporary files ?!? Was there really a single Google engineer involved in the process thinking "hmmm, this has certainly never been done before!".

Re:I'm Sorry, but...

[wren337]

This is a case of the USPTO saying "We don't understand this fully, we'll let the courts figure it out".

And the courts say "We don't understand this fully, we'll defer to the experts at the USPTO".

Re:I'm Sorry, but...

[Gr8Apes]

... I certainly don't know the ratio of good and bad patent grants. Alas as far as patents are concerned, Slashdot crown will accept nothing but impossible perfection.

In the realm of software patents, /. is pretty much "Patents == bad". Perfection would be NO patents.

Re:I'm Sorry, but...

[Anonymous Coward]

Well, how about you try to understand a new idea you haven't met before, written in patent-lawyerese, and find relevant prior art, which might be written in entirely different terms, from all existing technical literature, then formulate a justified formal decision, perform all the related bureaucracy, and leave enough time for the appeals process. You can't just say "no, that's obvious", you have to find similar already existing documented solutions, which may mean reading hundreds of pages of related documentation, most of which will turn out to be irrelevant after you compare all the details. I don't think I could do all that in a couple of days without compromising on quality.

Re:I'm Sorry, but...

[frosty_tsm]

Or they are normal people, without much domain knowledge, forced to handle too many cases in too little time, and fit within the rules of a broken system.

I personally find that to be the more plausible situation.

This is the picture I got from someone who works at USPTO.

Re:I'm Sorry, but...

[tambo]

> The USPTO is supposed to support itself with fees [uspto.gov]. The largest fee is for reexamination, creating a financial incentive to grant bad patents (which are likely to be reexamined).

That makes no sense when you look at the statistics. About 1,000 reexamination cases are filed every year. By contrast, the USPTO receives about 500,000 new patent applications every year. The total revenue from reexamination wouldn't even put a dent in the examination process.

Here's how it actually works. When you file a new patent application, you pay an examination fee. That examination fee gets you a little ways down the road (typically two office actions), and if the case isn't allowable by then, you pay another fee for a Request for Continued Examination, which gets you another two office actions. Etc. If you reach the point where the application is ready to be issued, you pay an issue fee, and you get your patent.

In other words - the USPTO funds itself by charging you every time it needs to do something for you, and the costs line up with the amount of work required by the PTO. It's exactly like a car mechanic, right? A mechanic has no interest in doing bad work now in the hope that you'll come back with more expensive work later. It just charges you, today, based on the service that you're asking for, today.

Re:I'm Sorry, but...

[tambo]

> Sure, but automatically deleting temporary files ?!?

Is every book entitled "Pirate Adventure" about the exact same story?

You can't just read the title - you have to read the claims. There's a whole lot more specific detail in the independent claims than "automatically deleting temporary files."

Re:I'm Sorry, but...

[Hatta]

Or they are normal people, without much domain knowledge, forced to handle too many cases in too little time, and fit within the rules of a broken system.

The right thing to do if you're given too many patents to examine is to grant none of them.

Re:I'm Sorry, but...

[jythie]

They are also severely overworked and undertrained. Patent review sounds simple from the outside, but we only tend to feel that way when we have personal knowledge of a particular patent. Throw a thousand various tech patents on any one of our desks and I would wager that even with our domain knowledge we would only recognize historical patterns a small fraction of the time.

Re:I'm Sorry, but...

[tqk]

The process should be turned on its head. You're applying for a patent? It's up to you to prove you deserve to get it, and if it's found that you can't prove it or you fudged the attempt, you'll pay a fine for the privilege. Any of the patent examiners discovering a bogus application wins a bounty! Why the !@#$ wasn't it designed to work this way in the beginning?!?

In a perfect world ...

Re:I'm Sorry, but...

[Anonymous Coward]

But the USPTO is populated by idiots.

Well, then, good news! We're hiring. If you'd like to put your money where your mouth is, and if you'd like to increase what you consider to be the average IQ of patent examiners, then go ahead and put in an application.

http://careers.uspto.gov/Pages/PEPositions/Jobs.aspx

Re:I'm Sorry, but...

[operagost]

It's pretty bad when you have to tell a Google engineer to "Google it".

Re:I'm Sorry, but...

[SuricouRaven]

The only problem is the size of the fine. Any fine large enough to be even a minor concern for a tech-giant like Apple or Microsoft would have to be larger than most people earn in a lifetime, and more than most smaller companies or individuals could risk. A fixed fine would be unworkable, and trying to scale fines to the income of the recipient can be a very complex task.

Re:I'm Sorry, but...

[SuricouRaven]

In which case you get fired, and replaced with someone who at least tried to take the job seriously.

Re:I'm Sorry, but...

[Anonymous Coward]

Until they decide to 'liquidify some of their IP portfolio to support short term opex' and a patent troll buys it

Re:I'm Sorry, but...

[Hatta]

Refusing to issue patents you cannot properly examine is the only way to take the job seriously.

Re:I'm Sorry, but...

[Anonymous Coward]

If they have too little time, they should summarily reject patents instead of granting them.

Re:I'm Sorry, but...

[tqk]

The process should be turned on its head. You're applying for a patent? It's up to you to prove you deserve to get it, and if it's found that you can't prove it or you fudged the attempt, you'll pay a fine for the privilege.

A fixed fine would be unworkable, and trying to scale fines to the income of the recipient can be a very complex task.

I don't think that's an insurmountable problem. You could make the fine == the filing fee, or even wave it based on an individual's circumstances (never filed before, you get a pass; second filing, filing fee; third filing, ...). Ramp it up for frequent filers/corps. Patent trolls would find it self-limiting. File a patent or attempt to extort fees on an egregiously bad patent that an examiner can easily invalidate, and you *lose* big time.

Re:I'm Sorry, but...

[Darinbob]

I don't think the Post Office deserves ridicule. It survives without funding from Congress and is self supporting. It doesn't make a profit but it's not supposed to. One of the big reasons it currently has financial problems was with Congress meddling and requiring them to pre-fund pension plans which no other government agencies have to do(not even its unions want this). Granted this is not the only financial headache. But this requirement from Congress really feels like an outright attempt to force privatization on the only agency that's self sufficient.

Re:I'm Sorry, but...

[surd1618]

Large organizations would respond by creating a new business with no capital or history to file each and every goddamn troll patent.

Re:I'm Sorry, but...

[Anonymous Coward]

> Sure, but automatically deleting temporary files ?!?

Is every book entitled "Pirate Adventure" about the exact same story?

You can't just read the title - you have to read the claims. There's a whole lot more specific detail in the independent claims than "automatically deleting temporary files."

Apparently, tambo, you are the one who needs to read the claims.

Re:I'm Sorry, but...

[jockm]

That is a policy of the patent office, who's rules are biased toward granting, than of the examiners.

Icon II

[skogula]

I used to service Icon II's which used a primitive form of QNX. When the hard drive filled up, it would start deleting old files. In a school, the oldest unmodified file was usually the master password file. Since these systems didn't have a built in root login, this means they were self bricking.

Re:Icon II

[Jeremy Erwin]

built in obsolescence,

This sounds like a submarine patent

[Jeremy Erwin]

A path name for a file system directory can be "C:temp\12-1-1999."

Why is google still using a C drive? And Temp is a bad place to store anything of value.

Re:This sounds like a submarine patent

[chrismcb]

The name of the patent is "Automatic deletion of temporary files ".... I'm thinking that MAYBE just MAYBE c:\temp is a great place to store temporary files. Mind you, I'm weird like that.

Re:This sounds like a submarine patent

[Jeremy Erwin]

I suppose it all hinges on which parts of the Operating system control C:/temp-- google's little cleanup daemon, or the host, which may not care all that much about google's fancy file system.

My new patent

[Air-conditioned cowh]

I have just patented a system of composing Slashdot posts where the attention span expires after......

this prior art should not matter

[TitusGroan8856]

this prior art should not matter, the patent should have been denied on the grounds that it is bloody obvious, that is a reason for rejecting them is it not?

Re:this prior art should not matter

[chrismcb]

Yeah, but just like the existence of prior art, the patent office doesn't seem to reject things based on that.

Remember when someone patented the wheel?

[ixarux]

No seriously.

http://www.hl7.org.au/docs/Australian%20Patent%202001100012.pdf

Not morons - just doing the job as intended

[dbIII]

It's just another version of a dog licence intended as a petty revenue stream. When a patent is granted that's proof of nothing other than the government is aware of it and has it on file - validity these days is apparently supposed to be sorted out in court and is none of the patent office's business.

Lack of research

[bkmoore]

Someone once said, most patent applications are a result of a lack of good literature/patent research.

Even older than that.

[Anonymous Coward]

I believe that In the 1960's Burroughs Corporation used a zero date for temporary files on their B-5500 systems running CANDE.

Employees looking for bonuses and what not

[Anonymous Coward]

Patents suck. Period.

Now that we've established that let me say employees are a good part of the problem. They are the ones answering the call from managers to "patent any 'innovation' they've created". I'm pretty sure lots of /. people working at big companies are complaining about software patents here and secretly working on submitting their dumb ideas to get a raise, bonus, whatever.

The things you

[Anonymous Coward]

get patents on in the US is just so retarded. I bet someone has patented green grass or wet water. How can people be this stupid? Oh wait, people pretend to be when there is money involved.

Crappy date format

[krigat]

I really wonder why a cutting-edge IT company like Google is using such a crappy date format in their patent example... it's not properly sortable and ambiguous. I know, it's just an example... still, it shows the obvious inexperience of whoever is responsible for filing that patent.Something like "C:\temp\1999-12-01\" would be the better way...

Come on Google, not even ISO8601?

[Barryke]

I would expect they use a YYYYMMDD notation or anything that flies with ISO 8601 .. but a M-D-YYYY format in ASCII inside a computer system, seriously Google?
What use does software have for that format besides communicate with Americans?

Re:Come on Google, not even ISO8601?

[SnowZero]

The example using a Windows drive letter was probably not added by an engineer, since the use of Windows in engineering is very close to zero. Regardless of who added it, it still serves its purpose as a short example, so there wasn't a reason to change it.

FWIW, Google's most common internal date format is:
      YYYYMMDDxHHMMSS
Where x is "s" for standard and "d" for daylight savings time. It sorts correctly even across DST shifts.

Of course, using dates for any reason other than informational hints is not considered a good idea -- time and date will vary over the globe at any given point. All data files will have internal records with unambiguous timestamps, but the name gives hints to someone doing maintenance work. If need be, it is also good enough for very coarse cleanup rules (delete stuff > 30 days).

Author is being intentionally stupid?

[Anonymous Coward]

There is a difference between (a) deleting a file on date X and (b) not deleting a file before date X.

Your math nerd card. Hand it in.

Patently obvious

[h8sg8s]

As the past head of the USPTO purportedly said, "Everything that can be invented has been invented." They should maybe take a que from him and actually research past patents for prior art. It's become less than a bad joke that you can patent anything these days as long as you make it sound new and obscure.

patent this

[Anonymous Coward]

you can also determinate separator using known 5/2 = 2.5 where second char is separator

The Problem Is That The Oversight of the USPTO...

[ios and web coder]

...is done by folks who have the technical knowledge (and people skills) of Tijuana pole dancers.

Tech companies know this, and have basically been daring each other to attempt patenting more and more outrageous things.

Sadly, with great success.

It's amazing what walking into a bar with lots of dollar bills will get you...

One un-arguably good thing about patents...

[DoctorBonzo]

is that they expire.

Maybe not fast enough for some of us, but too soon for others.

Or the summary is misleading propaganda

[raymorris]

The "summary" wholly misrepresents what the patent is about. It's not about having an expiration date in the filename at all. When someone advocating a position lies to me, as this submitter did, I figure the reason they are lying about the issue is because they realize that the truth doesn't support their position.

Rather than choosing an expiration date ahead of time, the patented method deletes a file (or not) based on multiplying the time to live by the inverse of the user's quota usage, plus the latest of several modification times. The patent covers only using that specific algorithm, and only when the TTL is represented within the filename.

Is that algorithm obvious? Several Slashdot commentors who say the are programmers read the explanation of the algorithm and still didn't understand it at all. One might say that if it's explained to you and you don't "get it", it's probably not obvious.

Re:Or the summary is misleading propaganda

[Maximum Prophet]

... Is that algorithm obvious? Several Slashdot commentors who say the are programmers read the explanation of the algorithm and still didn't understand it at all. One might say that if it's explained to you and you don't "get it", it's probably not obvious.

There's a fine line between clever and stupid. If an average programmer reads the explanation, and "Doesn't get it", it could be either. Most patents are very poor explanations for what they are about.

Re:Or the summary is misleading propaganda

[Anonymous Coward]

Being slashdot, I don't assume that the submitter is a liar, per se. There are certain topics (patents, copyrights, law) where slashdotters (except you and me!) are not just willfully ignorant but willfully retarded.

Re:Or the summary is misleading propaganda

[Anonymous Coward]

so they patented an algorithm.

Re:Or the summary is misleading propaganda

[tambo]

> When someone advocating a position lies to me, as this submitter did, I figure the reason they are lying about the issue is because they realize that the truth doesn't support their position.

I don't think it's flat-out lying. I think it's an example of the echo chamber effect.

The community believes that patents suck, that patent examiners are inept, and that patentees are using clever tricks to patent things that aren't new. So upon encountering any new patent, the submitters here don't do the hard work of reading the patent, parsing through the difficult claim language, and determining what it's all about. Instead, they read the title, maybe glance briefly at the abstract and the claims, and come up with a "basically, it's (something really simple)" summary, and post it as evidence of their beliefs about the patent system. A bunch of commenters then accept that summary without consideration, since it's yet another example of "bad patents," so they post a supporting rant about patents and increment their mental "bad patents I've seen recently" counter by one.

Of course, that process is flawed if the summary is an oversimplification of the claimed technique. Like this submitter concluding that the very specific technique presented in the independent claims is "basically, it's deleting temporary files," or "basically, it's deleting temporary files based on a modification date." But it's accepted without question because it supports the beliefs of the group. Hence, echo chamber.

Re:Or the summary is misleading propaganda

[tambo]

> There's a fine line between clever and stupid. If an average programmer reads the explanation, and "Doesn't get it", it could be either. Most patents are very poor explanations for what they are about.

But the "average programmers" here aren't motivated to try to understand it. They are motivated to find that the patent is worthless, because that's what the submitter wrote about it, and that's what they are predisposed to believe. So they are prone to glance at the application and say, "well, the claims have been mangled by lawyer-speak, but it's basically something about deleting temp files, which has been known since the 70's."

Re:Or the summary is misleading propaganda

[jythie]

Unfortunately, most commenters do not bother looking past 'patent references thing that was done before in its reference section, therefor it is patenting something that was already done!' and tend to leave out that if you make such references, the patent is supposed to explain why the new thing is substantially different.

Re:Or the summary is misleading propaganda

[Anonymous Coward]

Is that algorithm obvious?
yes

Re:Or the summary is misleading propaganda

[Anonymous Coward]

About 1/2 through I realized I couldn't think of anything that had implemented a system that met all the claims in the patent. It is clearly a detailed continuation of the far more simple temp file management systems already practiced.

Re:Or the summary is misleading propaganda

[Mycroft_VIII]

A big reason for patents is so new inventions are publicly documented so that when they expire they become part of the public domain so everyone can use them.
      If someone skilled in the field can't figure out the patent then it's a bad patent application and should be rejected until it's useable by others once it expires.

Mycroft

Almost

[Anonymous Coward]

find /tmp/* -atime +14 -exec rm {} \;

This way is better, relying on access time rather than modification time. It's quite plausible to have a file that isn't modified, but needs to be read often.

This is also how I handle dropboxes for transferring files via FTP or HTTP or sharing with a workgroup. Several folders e.g. 1Day. 3Day, 1Week, 2Week, 1Month... Drop your file into the folder appropriate for the desired period of time. A nightly cron job purges anything not accessed for the corresponding period of time.

Misunderstanding EXPDT

[Jay Maynard]

The original poster misunderstands what EXPDT= in JCL specifies. It doesn't mean the file will be automatically deleted after that date, only that the file may not be deleted before that date without the operator at the system console giving permission.

I'm no patent lawyer, but it would seem to me this is less prior art than he thinks it is.

Not so fast...

[MatrixCubed]

Blockly!

Can't fault Google for self defense

[walterbyrd]

As I write this, MS, Oracle, Apple, and others are ganging up on Google. The primary weapon is junk IP claims. You can read about it on Groklaw. Google is simply trying to fight fire with fire, against the worst scam artists in the industry.

Microsoft, BSA, Scott McNealy, others file amicus briefs in support of Oracle's appeal against Google ~pj

Wednesday, February 20 2013 @ 05:57 AM EST

Yesterday there were numerous amicus briefs filed all on the same day and all in support of Oracle against Google in Oracle's appeal at the Federal Circuit. None of the briefs are posted publicly yet, but they should be available soon.
Microsoft has filed one, together with EMC Corporation, and NetApp, Inc. Scott McNealy has filed one with Brian Sutphin. Can McNealy be a witness for Oracle at trial, which he was [PDF], and also file an amicus brief? Well, he has. The Picture Archive Council of America, Inc. has filed one with the Graphic Artists Guild. Also there's one from the BSA. And finally Eugene Spafford, Zhi Ding, and Lee A. Hollaar have filed an amicus in support of Oracle. Hollaar seems to file a lot of amicus briefs.

So why do these entities and individuals care about this Java API case, do you suppose?

never attribute to malice...

[mekkab]

... that which is adequately explained by stupidity. You give people too much credit and I'm sure the average joe at USPTO was being judged by how many applications s/he could get through; not get through competently. So, stupid managerial policy? Definitely the culprit.

/I realize there is a lot of assumption in that. Perhaps I'm also being stupid! :)

Cray had this also...

[klubar]

In the late 70's I worked on one of the original Crays and it had an option to specify a file deletion date (or retention time) when you created a file. The file would be automatically deleted (or maybe archived) at the appointed time. I've often thought that this would be useful in a desktop OS--when I create the file, specify that it should be deleted in 2 weeks. Same with email--it would be great if you could read an email and then indicate that its retention should be two weeks or one year... and then it would automatically disappear.

Reason for that!

[Runaway1956]

Think maybe there's a reason for that?

Like - maybe software might merit copyright protection, but never patent protection?

Re:Reason for that!

[Gr8Apes]

I'm not sure where you got the implication that I was against "no software patents"?

A sensible, profit-focused move

[frinkster]

Google has finally come to the realization that my browsing history from 2004 isn't worth the disk it is stored on.

taking a piss

[AndyKron]

When will the patent office issue a patent for unzipping and taking a piss? A piss I'd like to take all over the USPO BTW. FUCK THE GOVERNMENT!

Can someone explain this?

[Anonymous Coward]

Why on earth is Google referencing Windows/DOS path names? Shouldn't we see /var/tmp/12-1-1999/

No, the patent doesn't use an expire date

[raymorris]

I can't see a significant difference. In both cases they are using a metadata *expire* date.

No, the process patented by Google does not use an expire date in the metadata, so there's one big difference. In fact, the main utility of the patented process is that it does NOT pre-determine an arbitrary expiration date ahead of time. Read the abstract .

The problem with "a metadata *expire* date" as you describe is that I might do as I did just the other day - create a file which I intend to use for one day. Three days later, I'm still using it, so I don't want it deleted. With "a metadata *expire* date" it would have been deleted while I was using it.

The process patented by Google has time to live in the path, which is multiplied by the inverse of the quota usage and the result of that calculation is added to the greatest of several last modified times to calculate whether or not the file should be deleted now, based on the calculation of all three factors together. That way, files are not deleted just because back when they were created someone thought they wouldn't be needed for very long. Instead, they are deleted only if three things all all true - a) the file is marked as short term (TTL), b) the file hasn't been used in a while and c) they are running low on disk space. The patent covers te exact calculation used to combine those three factors.

Re:No, the patent doesn't use an expire date

[14erCleaner]

The patent covers te exact calculation used to combine those three factors.

Which begs the question: why is a patent necessary for this calculation? The answer, of course, is that patents are mostly created not to protect valuable trade secrets (which this isn't, by a long shot). Instead, they're used as currency to be traded with other large corporations. Patents have become like Bitcoins in a way; they're valued because of the effort it takes to create them, not because the result has any value.

Re:No, the patent doesn't use an expire date

[turbidostato]

"The patent covers te exact calculation used to combine those three factors."

So, in the end, the core of the patent is an algorithm... which are non-patentables to start with.

LMGTFY

[raymorris]

http://google.com?q=begging+the+question

Re:LMGTFY

[dzfoo]

Two can play this game.

http://oxforddictionaries.com/definition/english/beg?q=beg+the+question#beg__15

Check out definition #1.

Now, to avoid any further confusion in the future, make yourself acquainted with the following word: context.

          -dZ.

Re:Virtual memory, etc. was easy in the early 1980

[dzfoo]

>> companies that would grind you into crackers if given enough motive

Thanks for that. I can't wait to have an opportunity to use that phrase! :)

        -dZ.

RAISES the question. There's no begging.

[raymorris]

Saying "beg the question" in the incorrect, recent, sense doesn't even make any sense - there's no begging going on. It's just a matter of hearing "____ the question" and paying no attention the words one is parroting. I would agree that in most cases it doesn't much matter - it's just silly to completely ignore the meaning of the words you're saying.

Howeve,r you said "make yourself acquainted with the following word: context". Indeed, context matters. Let's consider context. If grandma says her old computer is "running out of memory" in the context of complaining about always needing to buy new stuff to replace old, it's okay if she really means "running out of hard drive space. In the context of a computer science discussion, you don't want to say "memory" when you mean storage space - it's important to be clear about what you mean.

The context, in this case, is the context of a debating an issue. That is precisely the context in which "begs the question" is a specific term of art. Misusing "begs the question" in the context of debating a policy issue is like misusing "running out memory" to mean drive space in the context of a computer science paper. Just as "memory" has a specific meaning in the context of computer science, "beg the question" has a specific meaning in the context of policy debate.

test

[Anonymous Coward]

please ignore this test comment

Re:The Problem Is That The Oversight of the USPTO.

[Anonymous Coward]

I would postulate that the average Tijuana pole dancer has much better people skills than your racist fat ass...

wait...what?

[tonytraductor]

So...any script I've written which writes data to a text file in /tmp , and then deletes that file (temporary), is violating a google patent? Like, even the ones I wrote 10 years ago?

Sigh. Another misleading Slashdot patent story

[cundare]

RTFM. Even claim 1, the broadest claim in the patent, is so narrow as to be credibly novel and nonobvious. I defy anyone to identify prior art that infringes on every element of this (or any dependent) claim in this patent. Holy jumpin jeezus, isn't there ANYONE at Slashdot knowledgeable enough to vet these "the patent system is broke, dude!" stories that appear every week? Seriously, Slashdot may need its own Amazing Randi.

That's why 'prior art' isn't patentable...

[Anonymous Coward]

But I guess it can be if the filer, lawyers, inspectors, and searchers were all born after the event! (I wasn't but I am none of the above ).

Do you think I could get a patent on a conductive something in a near vacuum or a noble gas atmosphere? Possibly a round thingie that might make it easier to push or pull things around that could be put under them?

Ban

[carys689]

Another reason why software patents should not be allowed. The obviousness of this "invention" is apparent. The USPTO is pathetic and Google is shameless.

Garbage Collection

[Stubbyfingers]

In LISP we used (gc) to round up the trash and delete it. On HP-3000 you could expressly define a file as ,TEMP and it would delete itself. SO, GOOGLE, Shit I was using in the 1980's before Sergey and Larry were driving cars is now patented by GOOGLE. I don't blame GOOGLE--I blame a borken patent system.

Re:Garbage Collection

[tmjva]

I too currently use an HP3000. It is a workhorse that keeps going and going and going ...

A file on the HP3000 can be defined in what is called the TEMP domain. It simply disappears when you log off. (Poof!)

It is an awesome computer by the way. It has never had a worm or virus (outside the lab, and the one it had, only affected a program on a dumb terminal.)

And of course HP still owns the rights. (AND the CODE and the BUILD procedures.) Runs the old languages. COBOL, FORTRAN, SPL, Pascal, BASIC, C, RPG, Perl.

Good times.

Re:Garbage Collection

[Stubbyfingers]

I know Tracy. This is why I still play Empire. :-) And occasionally WARP.

How about peer review of software patents?

[jcalcote]

Too bad we can't get the USPTO to use the software development community for open peer review of patent applications before they're granted. The field is too complicated for a handful of people to make all the decisions anyway. We'd get the sort of results we get from IETF or other standards organizations then.

Re:The Problem Is That The Oversight of the USPTO.

[ios and web coder]

:-*

I'll bet I know more folks with...interesting...backgrounds than you.

Show up, someday. W'll hang out with them.

Wait Until the HP Legal Dept. Reads This

[tmjva]

Oh yeah,

This is going to generate a few laughs on the few HP3000 forums.

Once the HP legal dept picks up on it, it is only going to get better.

Benefit of the doubt

[FreedomFirstThenPeac]

The USPTO does not guarantee that your patent cannot be challenged, it just gives everyone else a target, as in "Hey! I was doing that in 1971 and ...". The rest is left as an exercise to the destructive class, aka, "lawyers." (not to be confused with the Creative Class ").