Looks like the ID type of the row in the database was a signed integer, overflowed from ~2 billions.
In this case the problem would be a simple database structure design problem, using the wrong type (with heavy consequences).
On MySQL instead of int, using bigint would have allowed up to the theoretical value of 2^63-1 records.
Wow, you're quick! I'll bet *nobody* else here thought of that -here at slashdot of all places!
So you would rather have everyone wait for them to "approve" it in 3-5 years, tell people to start using it and then find out that there's problems with real-world use. At which point they have to go back to "debating" how to fix the problems, which might takes another 3-5 years.
The nature of standards is such that they're not standards until they're approved and recognized. Flaws within the standards result in further standards and modifications - but at least you can then guarantee a minimum baseline of support. That also includes flaws -- if it fails, it will fail the same way across platforms.
And we all see how android is filled with back doors and hemmoraging data. Moreover google is now back peddling and starting to lock things down. Sometime you want freedom sometime you want security. I'll take freedom on my desktop and security on my phone. why? because in the future the phone will be my credit card and for that I want something close to trusted plat form computing.
You're making the assumption that iPhone apps don't give away your data. Is it a valid assumption? (If they use any kind of ad network or app usage aggregator, then the answer is "no". )
Erm -- only eight other people have suggested that so far. My question is: why doesn't any one of you get out there and do it, and find the answer for yourselves?
If you're being a firewall which blocks _outgoing_ connections then you're in some environment where you are not going to download any files whatsoever anyway, and you're probably better off using snail-mail.
Umm... wait, what? Perhaps you think that a proxy server somehow prevents downloading of files via http or ftp? Since downloading web pages is also downloading files (as you noted), I'm not sure why you might think that other types of files are somehow prevented.
People with proxy-only access to the outside world are not relevant when it comes to downloading files beyond HTML.
Okay, so this is clearly a troll attempt -- or your sarcasm is so blatantly obvious that it overloaded my detector and I missed it.
It's great to talk about helping, but as soon as helping out inconveniences people (such as by costing money or affecting other network traffic) the generosity of people takes a sharp downward curve. You have only to look at any 'donationware' software project to see how effective it is to nicely ask people to help.
All a checksum can do in this context is make sure you have a file that matches the source. While this makes sure someone can't poison the data they're sharing, it does *not* make it any safer than a standard download - it just ensures that it's not inherently *less* safe. You still have no guarantees as to the reliability of any particular source - that risk applies to both direct and torrent downloads.
Not to mention the nasty tendency that bittorrent has to saturate consumer-grade routers in a nasty way, even when not using all available bandwidth. I suspect ISP routers wouldn't handle it much better if it suddenly became the defacto standard method of transferring files.
That's what TFS said, though I could find nothing to substantiate it -- so I take it with a grain of salt.
But aside from that, my post is in the context of what the OP was discussing - if they hadn't signed the agreement to begin with. Further - it's still a moot point, as there are no OSS options out there for this that I know of.
But really, does this mean I have to now own proprietary applications to view public documents? Thanks for so much you've left us Pawlenty.
Well no - remember, Open Office or whatever it's called now claims it can read and write MS formats with no problems. So I don't see what your concern is...;)
Cost in manpower required to replace every windows server with an OSS server, including configuration for like functions - using staff nnot familiar with the environmetn?
Cost in hours to wipe all desktops and install OSS OSes?
Cost in hours to replace every piece of software that has no counterpart with a custom written version of the same?
Cost in hours to find a way to get all those stupid custom Microsoft Access applications working on a completely new platform?
Cost in hours to fix all of the intranet web sites that were built to run in IE6 only?
Cost in hours to correct formatting issues in office documents that don't display correctly in their OSS counterparts?
Cost in helpdesk support hours for the steep learning curve many users will face?
The equation is often not as simple as "well it doesn't have a licensing cost". While I think that MS"s ongoing "tco' campaign was a bunch of crap, I also think that to dismiss the actual costs associated with a massive conversion of this nature does more harm than good - because if someone buys into it and encounters the above (and the hundreds of smaller issues I didn't list), then it's OSS that comes out looking bad. Presenting a realistic (and honest) perspective lets people make the choice with their eyes open - knowing that if they get past the initial pain, it *does* get significantly cheaper on an ongoing basis.
The thing the article doesn't tell you in detail is that the agreement precludes the use of open source software, which could have saved the taxpayers millions of dollars. And once such a large organization goes Microsoft, it's difficult to go back. Isn't it interesting that these developments occur right before elections, as senior officials are trying to keep their jobs with a new incoming administration? What do you think,
I think this is classic flamebait. First: it's seldom as simple as saying "Hey let's replace everything with OSS!", because unfortunately there *is* at minimum a manpower cost when switching all of your existing systems over. There may be a manpower cost here - but if Microsoft is providing tools to make that transition simple, cheap, and/or free , then they're already a step ahead of their open source counterparts.
In addition, what open source options are there for hosted services of this type and scale? I'm not aware of any (but could be wrong). On that basis alone, the fact that they didn't go OSS is completely irrelevant because there are no OSS offerings in the running to begin with. And I suspect that if/when such services come into being, they will cost similar to what MS is charging, because now it's not about the software licensing - it's about the services and convenience of not having an IT staff to do the maintenance work.
Finally, who really thinks so narrowly as to believe that the software platform plays any role at all in whether senior officials "keep their jobs with a new incoming administration"? Software platform is a background part of these people's day-to-day activity - it's a tool that should do its job and stay out of the way. For the administration it's just one more cost in the budget checklist - and probably not a very significant one, relative to all of the other spending a state does on a daily basis.
Unless/until Windows servers are free and as effective at what I need as Linux servers are, I'll be running Linux servers for home and business. For my needs, it's the best solution - changing over would cost me significant money that I had not planned to spend. But for the state of Minnesota, it's awfully hard to evaluate whether or not that holds true on the basis of a PR release, a computerworld writeup, and a loaded slashdot summary.
Why can't submissions provide actual sources? In this particular case, we got a link to a blog - which linked to another blog - which linked to a techcrunch article - which linked to another techcrunch article - which linked to a dev mailing list. Would it have been so hard to provide the direct link to at least the techcrunch article which provided far more details than the random blog analysis of the same?
When you incorporate advertising "services" and usage statistic tools into your apps, this is what happens. You get the convenience of "free" tools which make your life easier; either by automatically handling ad imprints (and earning you some money); or by providing you with app usage statistics -- or both.
On the surface you don't pay anything for these tools. They integrate nicely into your app, and you only have to add a few lines of code -- the essence of what good developer's tool should provide. But it's free to you only because you passed the cost along to your users - often without realizing it. In exchange for the convenience provided for you, you've decided that your users' information, attention, viewing habits, and even privacy are fair currency with which to pay for that service.
If you value your customers, do the research before blindly incorporating these "free" tools into your applications.
As you've started ignoring the points you had no answer for, there's probably not much point in continuing. I do apologize for the ad hominem though. I made an assumption based on how you phrased your statement -- my bad.
You spun his statement and injected the idea that he said apps are necessary. He did not say that. You did.
he said:
Otherwise, expect us to live our lives by any means necessary.
In the context of a discussion about pirating apps. So how did I inject this idea, given the context of the discussion?
But there are certainly apps necessary... as in todays world, you better know a few of them if you want a job that isnt already taken by an illegal mexican.
I don't see anything that says they do intend to stop it - only that they don't want them. App World approval currently consists of them validating that your app does what it says it will, on the phone/software versions you say it will.
Slashdot Mirror
Bah. You went and gave a serious reply to my snide comment, and now you've ruined my fun. I'm taking my toys and going home. ;)
Looks like the ID type of the row in the database was a signed integer, overflowed from ~2 billions. In this case the problem would be a simple database structure design problem, using the wrong type (with heavy consequences). On MySQL instead of int, using bigint would have allowed up to the theoretical value of 2^63-1 records.
Wow, you're quick! I'll bet *nobody* else here thought of that -here at slashdot of all places!
The point is that a standard can exist even if it's 100% unimplementable - as long as the standards body votes it that way.
Actually, you're combining a few different issues. Naturally there can't be a 100% official standard until it's bug-free and proven implementable.
Tell that to Microsoft and their officially standardized Office format ;) Seriously, wh
The difference here is that all major players indicate willingness to be compliant with HTML5 - but can't do that 100% until it's official.
Never would have realized - thanks! We'll get right on that!
Sincerely,
The Rest of the World
So you would rather have everyone wait for them to "approve" it in 3-5 years, tell people to start using it and then find out that there's problems with real-world use. At which point they have to go back to "debating" how to fix the problems, which might takes another 3-5 years.
The nature of standards is such that they're not standards until they're approved and recognized. Flaws within the standards result in further standards and modifications - but at least you can then guarantee a minimum baseline of support. That also includes flaws -- if it fails, it will fail the same way across platforms.
And we all see how android is filled with back doors and hemmoraging data. Moreover google is now back peddling and starting to lock things down. Sometime you want freedom sometime you want security. I'll take freedom on my desktop and security on my phone. why? because in the future the phone will be my credit card and for that I want something close to trusted plat form computing.
You're making the assumption that iPhone apps don't give away your data. Is it a valid assumption? (If they use any kind of ad network or app usage aggregator, then the answer is "no". )
Do they let you keep the piece of your brain they removed in an iJar?
They tried, but then Oracle sued.
Dude... I'm counting my heartbeats now. ...
I think the gap between the last two was a little slower.
I hate you.
Erm -- only eight other people have suggested that so far. My question is: why doesn't any one of you get out there and do it, and find the answer for yourselves?
If you're being a firewall which blocks _outgoing_ connections then you're in some environment where you are not going to download any files whatsoever anyway, and you're probably better off using snail-mail.
Umm... wait, what? Perhaps you think that a proxy server somehow prevents downloading of files via http or ftp? Since downloading web pages is also downloading files (as you noted), I'm not sure why you might think that other types of files are somehow prevented.
People with proxy-only access to the outside world are not relevant when it comes to downloading files beyond HTML.
Okay, so this is clearly a troll attempt -- or your sarcasm is so blatantly obvious that it overloaded my detector and I missed it.
It's great to talk about helping, but as soon as helping out inconveniences people (such as by costing money or affecting other network traffic) the generosity of people takes a sharp downward curve. You have only to look at any 'donationware' software project to see how effective it is to nicely ask people to help.
All a checksum can do in this context is make sure you have a file that matches the source. While this makes sure someone can't poison the data they're sharing, it does *not* make it any safer than a standard download - it just ensures that it's not inherently *less* safe. You still have no guarantees as to the reliability of any particular source - that risk applies to both direct and torrent downloads.
Not to mention the nasty tendency that bittorrent has to saturate consumer-grade routers in a nasty way, even when not using all available bandwidth. I suspect ISP routers wouldn't handle it much better if it suddenly became the defacto standard method of transferring files.
But aside from that, my post is in the context of what the OP was discussing - if they hadn't signed the agreement to begin with. Further - it's still a moot point, as there are no OSS options out there for this that I know of.
But really, does this mean I have to now own proprietary applications to view public documents? Thanks for so much you've left us Pawlenty.
Well no - remember, Open Office or whatever it's called now claims it can read and write MS formats with no problems. So I don't see what your concern is... ;)
Cost in manpower required to replace every windows server with an OSS server, including configuration for like functions - using staff nnot familiar with the environmetn?
Cost in hours to wipe all desktops and install OSS OSes?
Cost in hours to replace every piece of software that has no counterpart with a custom written version of the same?
Cost in hours to find a way to get all those stupid custom Microsoft Access applications working on a completely new platform?
Cost in hours to fix all of the intranet web sites that were built to run in IE6 only?
Cost in hours to correct formatting issues in office documents that don't display correctly in their OSS counterparts?
Cost in helpdesk support hours for the steep learning curve many users will face?
The equation is often not as simple as "well it doesn't have a licensing cost". While I think that MS"s ongoing "tco' campaign was a bunch of crap, I also think that to dismiss the actual costs associated with a massive conversion of this nature does more harm than good - because if someone buys into it and encounters the above (and the hundreds of smaller issues I didn't list), then it's OSS that comes out looking bad. Presenting a realistic (and honest) perspective lets people make the choice with their eyes open - knowing that if they get past the initial pain, it *does* get significantly cheaper on an ongoing basis.
The thing the article doesn't tell you in detail is that the agreement precludes the use of open source software, which could have saved the taxpayers millions of dollars. And once such a large organization goes Microsoft, it's difficult to go back. Isn't it interesting that these developments occur right before elections, as senior officials are trying to keep their jobs with a new incoming administration? What do you think,
I think this is classic flamebait. First: it's seldom as simple as saying "Hey let's replace everything with OSS!", because unfortunately there *is* at minimum a manpower cost when switching all of your existing systems over. There may be a manpower cost here - but if Microsoft is providing tools to make that transition simple, cheap, and/or free , then they're already a step ahead of their open source counterparts.
In addition, what open source options are there for hosted services of this type and scale? I'm not aware of any (but could be wrong). On that basis alone, the fact that they didn't go OSS is completely irrelevant because there are no OSS offerings in the running to begin with. And I suspect that if/when such services come into being, they will cost similar to what MS is charging, because now it's not about the software licensing - it's about the services and convenience of not having an IT staff to do the maintenance work.
Finally, who really thinks so narrowly as to believe that the software platform plays any role at all in whether senior officials "keep their jobs with a new incoming administration"? Software platform is a background part of these people's day-to-day activity - it's a tool that should do its job and stay out of the way. For the administration it's just one more cost in the budget checklist - and probably not a very significant one, relative to all of the other spending a state does on a daily basis.
Unless/until Windows servers are free and as effective at what I need as Linux servers are, I'll be running Linux servers for home and business. For my needs, it's the best solution - changing over would cost me significant money that I had not planned to spend. But for the state of Minnesota, it's awfully hard to evaluate whether or not that holds true on the basis of a PR release, a computerworld writeup, and a loaded slashdot summary.
As for "helping the stimulus", imagine how bad the US would be without all those "jobs saved"... :-P
I'm imagining how bad the US will be in another decade, because of those "jobs saved" and other spending.
Why can't submissions provide actual sources? In this particular case, we got a link to a blog - which linked to another blog - which linked to a techcrunch article - which linked to another techcrunch article - which linked to a dev mailing list. Would it have been so hard to provide the direct link to at least the techcrunch article which provided far more details than the random blog analysis of the same?
On the surface you don't pay anything for these tools. They integrate nicely into your app, and you only have to add a few lines of code -- the essence of what good developer's tool should provide. But it's free to you only because you passed the cost along to your users - often without realizing it. In exchange for the convenience provided for you, you've decided that your users' information, attention, viewing habits, and even privacy are fair currency with which to pay for that service.
If you value your customers, do the research before blindly incorporating these "free" tools into your applications.
As you've started ignoring the points you had no answer for, there's probably not much point in continuing. I do apologize for the ad hominem though. I made an assumption based on how you phrased your statement -- my bad.
You spun his statement and injected the idea that he said apps are necessary. He did not say that. You did.
he said:
Otherwise, expect us to live our lives by any means necessary.
In the context of a discussion about pirating apps. So how did I inject this idea, given the context of the discussion?
But there are certainly apps necessary... as in todays world, you better know a few of them if you want a job that isnt already taken by an illegal mexican.
Clarify?
I don't see anything that says they do intend to stop it - only that they don't want them. App World approval currently consists of them validating that your app does what it says it will, on the phone/software versions you say it will.