Slashdot Mirror
Help Write An Open Data Format Bill
AdamBa writes "There has been a lot of discussion of open source bills, but I think open data format bills have a much greater chance of actually becoming law. Over at the Open Data Format Initiative site, I have written an article explaining "Why Open Data Format Laws Are Better Than Open Source Laws". I also have a sample Open Data Format bill; I invite comments from slashdot readers, in particular on how the sample bill could be improved."
...might or might not now help linux ... a lot of people I've talked to just don't like openoffice (and I've noticed that big spreadsheets are intolerably slow). There's a bit more work to be done there, besides the standard ms office problems. (That said, it wouldn't hurt a bit). But it would definately help a company like apple.
Imagine flawless powerpoint support in keynote.
You say
When it hits the legislature where it will quickly be rewritten to allow MS to do whatever it wants.
The author has the right idea; it would certainly make building compatible third-party implementations of data-processing software (by that I mean nearly all software) much easier than if the file formats were closed, as they are now.
The trouble is that the distinction between code and data isn't as bright a line as you might think. I don't mean code/data segmenting; if you think about what evals and ELF are, you know what I mean.
Algorithms (not to mention software in the US) are certainly patentable -- and efficient data storage mechanisms are too. Think about what a gold mine in terms of IP value a hash table would have been if it invented at a commercial organization instead of in academia.
So, the conundrum we still face is that there are still valid IP claims WRT data structures, because oftentimes, as much thought is put into them as into the processing software that reads and writes them.
On the second bullet you say "computer data owned by the [government] be permanently available to the [government] throughout its useful life."
Somewhere you may want to define "useful life", I could see this as a possible loophole, this term could mean different time lengths to different people.
Stuff this open data format crap.
I think data formats are patentable... because some of them are inventive and require real ingenuity. eg: video streaming, etc.
If you don't award patents, you don't get the great inventions. (cf: Edison)
from,
your friendly patent attorney
Goverment? What goverment? We are talking about open data formats for Internet, not for national network of one country, right? If so then the law should be international. Otherwise - it's wasting of time and efforts.
Less is more !
"A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
I don't think people want open source "laws." We want freedom of choice.
We do not want to be locked into a particular software package so they can exthort money out of us. Charging for what amounts to an upgrade is just wrong. Charging for and upgrade that makes the software do what you said it would do in the first place is wronger. Charging for security fixes is wrong. I don't like to see my tax dollars wasted. If you are a private company, that I have no stock in, do what ever you want...
I have never seen Ford charge their customer for a recall...Only in software.
With that being said, common sense should dictate to anyone in government who is in charge of purchasing that an open data format is much more inviting than a non-open format. But I can see that just because this should be common sense does not neccescarily mean that it is.
Here is my only concern: I don't like to have the government's hands tied, one way or another. Sometimes the market does not produce a product to satisfy these conditions or the market will produce an open data format product and a closed data format product where the closed data format product is superior in all ways but the obvious one. Now, I want the government to be able to evaluate their needs (considering how likely it is that they will need backups of the data - for instance if the data is only going to be temporary work product for an internal office and will eventually be converted to HTML) and make an informed decision. But I suppose in many cases it is better to lay down a rule which will work positively in 99% of all situations.
I hate liberals. If you are a liberal, do not reply.
What ever happend to independent and competing ideas that will eventually succumb to the better format?
...
Remember LZH, ARC, PAK, ARJ, JAR, ZIP, RAR, ACE, ZOO,
All competing algorithms and eventually ZIP won [or more so deflate] for the most part and BZIP2 to a certain degree...
If you stiffle independent thought before you get out the door why bother?
Tom
Someday, I'll have a real sig.
I would prefer a complete free and open standard with no patent encumberment, but a second-best option would be to have an open and fully documented standard that may or may not be subject to patents, along the lines of the stuff put out by the MPEG consortium (MP3, AAC, etc. are all patented, but are also completely open formats). This way at the very least you have an guarantee that your data won't be permanently inaccessible, because when the patent expires you can reimplement the open specification.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Except the author thinks that software companies are just going to lovingly hand over specifications on their file formats. Throughout computing history, software companies have used file formats to lock people into their product, force them to buy upgrades, and use them as leverage for strategic partnerships. Ie- it's a huge cash cow to have proprietary formats.
The author, in the very first paragraph, dismisses "open source laws", saying they "won't work". Huh? Says who? Then he says his WILL work. You can't just make enormous blanket statements like that without backing them up!
In his second point, he says "Many open source laws seem designed to force a government to replace Windows/IIS/Office with Linux/Apache/*Office". Where is he getting this crap from? He goes on to dismiss the security benefits of open-source software, the cost savings...he pretty much dismisses every single argument for considering open-source software, with nothing to back up his reasoning. He 'thinks', therefore it is.
My impression of "open source laws" was that they instruct government agencies to -CONSIDER- open source solutions- it does not FORCE them to use them- yet he makes it out like there's a cop sitting there with a gun to a government IT manager's head, saying "Go ahead punk. Make my day. Install Office. Are you feelin' lucky, punk?" Maybe I misread it, but his idea seems to be to -FORCE- companies to release file formats if their software is used by the government.
One requires you consider open-source software, leaving commercial software companies plenty of oppertunity to compete if they've got a better solution(remember, open-source is not $0, you still have labor involved, possibly a migration, maybe staff training and hiring, maybe different equipment.) Open source doesn't automatically, if ever, "win" outright simply because it's open-source, yet the author seems to think open-source always will, and hence open-source laws will be bad because there will be a huge inconvenience to the government or the software companies. Again, consider- not force! If the manager thinks commercial software will overall be better, he/she will make that call.
The other forces software companies to do something that threatens, in a BIG way, how they compete against other companies.
Now, which has more of a chance of failing?
Please help metamoderate.
Admittedly i read the 'bill' a bit quickly, but there is one problem (which seems to be partially addressed later on) is that all data be in human readable format. Well this is good for string data. However I very much suspect there are going to be number formats that are much better stored in their binary format. Now as I recall from college COBOL's strategy as to have everything stored as string data with (interesting) defining headers. As I also recall this was to help portable between COBOL and other foreign (computer wise) data retrieval systems, like when data was passed around on tapes.
This is all well and good, but you trade off portability for parsing overhead. Of course CPU time is way cheap as to compared to yesteryear. However you will end up with people's nutjob parsing routines, I mean hell CPUs and calculator software still get shipped making math mistakes. Reading in data in any of the standard int and float storage formats is cheaper and you don't have to worry about joe-bob desk jockey hosing up an import routing and your tax bill being 100x too large.
Now as most of us who have been thru a proper programming class know there are routines for these kinda things.. but again beware of joe-bob who wants to do it 'better' or whatever. Perhaps along with open data one needs to define a data import stadard?
Open data formats are a good start, but don't fool yourself into believing they are sufficient. HTML is perfectly open, but how many web sites are there that test against anything but IE? How many rely on the behavior (even the bugs) of IE?
The non-openness of Microsoft Word's document formats is not the biggest obstacle. There are plenty of office suites now that can read them. The biggest obstacle is that people write their documents expecting them to look exactly as they do in Word. That's not something that open data formats by themselves will solve. Microsoft could move entirely to XML, and that problem would still remain.
1 - say a wordprocessor developer decides to get rid of the concept entirely of files - does that mean they have to disclose how they store the information? ie: perhaps using their own disc format and writing sectors/tracks directly?
2 - where does the difference in data and program exist? ok, most of the time it is plain and clear, but what if the text of a wordprocessor wasn't stored as text, but rather a program that was interpretted to give the text layout? Also, is a BASIC program tokenised not data for the BASIC interpreter?
3 - Why should a programmer document how all their data is? They may not even know, it may be purely left to how the language that they used writes their data structures. Some of the data may be specifically not designed for 'users' to exchange? ie: indexes of a database.
4 - forcing people to disclose how their data is - is just not right. A programmer should have the right not to disclose how their program works.
Julian
This is like saying ASCII works for everyone. It is more than just the format on which the "real format" is based on. It would be ideal to have a full specification or detailed, non obfuscated documentation.
S
1) well then said app wont be running on any current OS I can think of. Reimplementing low level data storage routines is plain stupid, especially for a high level app like a word processor. (i can hear the grumbling 'but that was just an example you are too literal.. etc)
2) This argument is older than dirt and way too contrived. If a user is inputting records into a database its not code. Besides interpreted basic (as its a structured high level programming language the 'data' is already defined. Besides if the govt. is writing apps in interpreted basic you better ph33r f0r y0ur l1f3!
3) This is the whole point of the exercise, to write your data on purpose such that you document its output. In addition indexes to a database *could* be made user friendly, but since they are ephemeral anyway and are rebuilt and updated as needed and not generally data that is input by a user for long term data storage its kidna falls outside of the bounds here.
4) Well this is forcing the govt. to document their data structures that contain data that is not internal to the govt. And since ostensibly we all own the dat that the govt. has we can tell them (ideally) to do whatever we want.
Instead of making major corporations, such as Microsoft, open their formats, perhaps Word and similar programs should be required to have the ability to read and/or write major open formats, such as StarOffice/OpenOffice formats.
Yes, I could see arguments along the lines of "but we can't possibly support every format," but in the case of behemoth corporations, it should be doable for at least the major formats. This could also be said to stifle competition, but in Microsoft's case, it has already done enough of this, and this measure could actually restore competition.
I've read Grocklaw. BoycottNovell, you're no Grocklaw
You're too literal -- Within a few years MS Office will be transformed into a server-based content management system. There won't necessarily be any "files", just database content that is rendered dynamically. This nicely eliminates both the DOC file issue and the piracy issue.
there is no well defined distinction between source code and data, nor is such a definition possible (think Lisp).
.Net runtime).
if closed source is legal, but closed data formats are not, MS will just make their data documents "runnable" (think embedded VBscript macros, etc...) and say "that's not just your average data document; it's a *program* (value added for customer satisfaction!), and it's *compiled* (ie obfuscated) for performance; therefore, we don't have to make it open."
MSword and its ilk will just become an interpreter for "running" your saved documents (think C#
No, it was about choice. He was saying you should not be forced to do something that is common sense; that's a debate about choice.
I'm going to ignore most of your post on the grounds that you either have deliberately missed my point, used your own strawman arguments, or are just thick (I'll let you decide which...). However, there's a key point I want to address:
So you're saying the government is allowed to (pre-)determine for me a choice that could affect my living or dying? Do you think they should be able to do this in every case? What if I am going to have surgery where I have a 10% chance that I could die or be severely maimed? Because it's "only a 10% chance" does that mean the government gets to decide for me?
Actually, in general, yes I do believe that.
Most modern societies believe that we have a duty of care to other members of our societies. This results in such things as publicly funded hospitals.
Such facilities are almost always heavily utilised. In such situations, most societies look at ways of reducing the need for these services.
The seatbelt case is an excellent example. When seatbelts first started being available in cars, some bright spark noticed that there were relatively fewer injuries amongst seatbelt wearers. Statistical studies were done and they showed that, on the whole, wearing a seatbelt meant you were less likely to be hurt. Local governments started passing seatbelt laws. Other studies were done showing a large correlation between seatbelt laws and reduced hospital demand. So everybody picked it up.
If seatbelts weren't required, hospitals would have even greater demands on them. Hence the law. I feel that governments have the perfect right to curtail behaviour that puts higher demand on government services.
So, that's how a private choice (the decision to wear or not wear a seatbelt) becomes something worth legislating.
This is the dumbest argument. Those people who die or are severly hurt by seat belts should not have been legally bound to wear them. Perhaps they would've worn them anyway, perhaps not, but at least it would've been their choice and not the government's.
First obvious point: It's always their choice. The car doesn't have a sensor in it to not let it start if the seatbelt isn't worn. They can drive without a seatbelt. It's just now, in addition to risking greater injury in the event of an accident, they also risk getting fined.
Second obvious point: Yes, seatbelts sometimes result in greater harm than they do good. Laws are aimed at the benefit of society as a whole, not individuals. On the whole, society is helped by seatbelt laws.
To take another example, some people think vaccinations should be a personal choice. Vaccinations occasionally have undesirable side effects. Despite that, high vaccination rates for measles, polio, chickenpox, whooping cough, dyptheria, and other nasty diseases have vastly reduced breakouts and associated fatalities. However, it's also been shown that when vaccination rates drop below a threshold (like, 80%), the breakouts start occuring again. Because of this, if a government wants to make vaccinations mandatory, then I think they have every right to.
The desire to have governments not intrude on these apparently individual (and often common sense) choices is, at heart, an entirely selfish one. It ignores the fact that these laws are not designed to curtail your rights, such as they are; they are designed to help society as a whole.
And to tie it back to the topic: a government body mandating that suppliers of software to it use a publicised data format is a good thing. It may be common sense ("Of course I want the ability to access my data in any way I want"), but that doesn't mean it's not something that should be legislated.
"Software is too expensive to build cheaply"
The first approach takes too simplistic a view of the type of software that governments use. Much of the it is customized for specific tasks such as processing drivers' licenses, and the market for providers of such software is presumably small. If software vendors release their software as open source, they may find that cash-strapped governments in other states gladly help themselves to it for free, so the vendor may get only one paying contract instead of fifty. Therefore, it's quite possible that governments won't be able to find companies willing to provide them with open source software, and then what alternative do they have?
I just covered this recently in an op-ed piece in the business journal. Let me make this simple: governments do not exist to provide business opportunities to software developers. Reread that until it makes sense.
We (collectively "the people") shouldn't have to pay twice for a piece of software. The open source world needs to come up with solutions to boring problems, like drivers' license registration. If a particular company doesn't want to be a part of that, then so be it.
Michael
Do you have ESP?
The problem in your argument is that, despite what the ads say, seatbelts don't save lives.
Yes, that's right. Your chances of death in a car accident (at least, death without getting to a hospital first) are not decreased significantly by wearing a seatbelt. Seatbelts prevent the "being tossed around the car, suffering massive bruising, broken bones, spinal damage and internal bleeding" type of injuries. Most quick deaths from car accidents occur due to broken necks from whiplash (seat belts don't help, may hinder), or foreign object intrusion.
So how are your feelings of someone risking greater, and more costly, injury, but without raising their chances of cheaply dying very much?
(Nice pragmatic attitude, BTW... reflects mine a lot)
"Software is too expensive to build cheaply"
If software vendors release their software as open source, they may find that cash-strapped governments in other states gladly help themselves to it for free, so the vendor may get only one paying contract instead of fifty.
IIRC, the GPL specifies that you only need to make the source available to the entities to which you distribute binaries. I.e., if Michigan gets a contract with some company for some software, the company can release the software as GPL and only provide the source to the Michigan state government. It is then in the best interests of the state government to NOT redistribute the software. That way another state, say New York, also buys the software to make sure that the company is still there next year. The best thing is that if company X pisses off the state, they can take the source and hire company Y to maintain it. If I understand the intent of the GPL, this is the kind of choice that is available with free software.
Compare that to a previous post (I don't recall the exact thread) where the poster pointed out that his organization used a particalur personnel management product. PeopleSoft bought the competitor and disconitnued the product. Security flaws were found, his company had to spend $2M to switch to PeopleSoft. Had that software been GPL, they could have just hired another company to fix/maintain it. Again, in that situation it is in the company's best interest to NOT freely (as in beer) redistribute the program.
I think the same can be said for many non-commodity software products. I mean, how many people actually need or want (or have the hardware) to run a PeopleSoft level program at home? Better yet, if a company pays, say $250000, for a program and accompanying support, what incentive do they have to turn around and give it away? It seems to me that the GPL is really perfect for situations like this.
Wrong. By making the native format of the application open the format must fully represent the data. With input/export formats it's just too easy for unethical companies to leave out some critical element, making them useless for interoperability. eg. M$Word's useless "XML" format.
I've seen all the comments on loopholes and inconsistencies in this proposed law. This is why the politicians and corporations normally write the laws.
Unlike us, they know how to eliminate loopholes and gross internal inconsistencies. Take, respectively, the tax laws and the DMCA for example.
.. you lose a freedom. Let the market decide the formats. This law sounds like something Europe would make.
If you have a law forcing data to be in properly-documented formats, proprietary software companies are free to implement apps that work with that format, without open-sourcing anything.
.doc format works".
They'll just have to compete (with open-source apps, and with other proprietary apps) on features, ease of use, reliability and things like that, that actually help the user, as opposed to competing on "but we're the only people who *really* know how
(IMO governments should use open formats; private companies shouldn't be required to use open formats, except when they interact with governments, which would hopefully mean they see the sense of going with open formats throughout.)
(Of course, if someone decides the best way to document their format is to open-source a sample implementation, that'd be nice too...)