Slashdot Mirror
Should The Government Go Open Source?
The Village Voice is running a story about New York City's troubles with the vendor who built the subway Metrocard system - magnetic swipe cards that work so well, I almost got arrested once because the system was... deficient. Though the story is about a specific situation, the general problem (municipalities becoming captive to corporations with specialized expertise) is extremely common. (And governments spend a fortune on such contracts.) The author-recommended solution is that the municipalities develop communal, reusable systems. I can imagine plenty of systems that would benefit - start with the air traffic control system. Is this the future of government-developed code? Or will it continue to be one-off, hideously expensive, closed code?
This New York story sounds really bizarre to me. It smells of ... corruption. Here, the transit authority has failed to all the standard due diligence required for public contracts. At least, compared to the usual pratice in Europe. Or may be this situation is the norm in the US, and that would be scary.
I'm mostly aware of the French practice, Germany is quite similar yet GB is much closer to the US and very often completly screws up. European practice is not to open source, not yet, but to require that any vendor must give away to the public administration, at some measures, the IP of any system developed on purpose for this administration. Full access to the code, full documentation and the right to transfer the existing IP to another vendor to pick up and maintain the existing system.
Of course, the vendors have some guarantees they won't get thrown away on a wink and the ability to transfer something to another vendor is somewhat related to a breach of contract from the orginal vendor. But the definition of a breach of contract is heavily skewed towards the client and also the contract are always time-bounded.
Now, that's the theory. The reality is that for big systems, they are not so many vendors, that they don't like to tramp on each other play ground and that it's actually easier to recommission the original vendor. On the other hand, the State is something like 45% to 55% of the GDP, public contracts are a huge part of the market, and in public competitive bidding, the price is not the only criteria. The best bidder can be shoved aside if it has proven ifself unreliable on prior contracts. One really has to screw up big time to piss the State administration but once you're there, you're really 6 feet deep in shit.
Also, specifically for France (I don't know about Germany and GB again is completly different), there's an additional trick. Litigation with a public or para-public administration (for instance, Social Security) are not brought to the normal civil courts but to the so-called administrative court. It's not the same law that apply to mere mortals and to the State. Guess what ? You really need to have a good case against the State if you want to prevail... That's a great incentive to play nice.
I'm an aerospace research engineer at NASA Ames. We have a program here called the Employee Suggestion Program, and I recently suggested that Ames dump MS Office and use StarOffice instead. Actually, I don't use a word processor at all (I use LaTeX, HTML, and PDF), but it bothers me that the government pays so much money directly to MS when we can get the same functionality for free. Also, many engineers with Unix workstations also need a PC or a Mac just to run Office. It's ridiculous.
Russ P.
http://RussP.org
When the metrocards were introduced, people found that they could get free rides by bending or scratching the cards just so. So they upped the tolerence.
NT administration is still tricky hard work that requires a lot of specialized knowledge no matter what the four-color glossy literature tells you.
OK, now try to get one of those Expert NT or Unix admins to work in an underfunded laboratory and spend 90% of their time doing routine admin work like setting permissions. It's not going to happen.
NT admin is a specialized task, but the product is hacked in such a way that small networks can be setup by people who don't understand it (such as MCSEs) and it sorta works. You don't need someone who really understands it until you get to an order of magnitude larger network.
On the other hand, Unix admin is harder to fake. You either know the magic commands, or you don't.
Mission critical systems that could result in the loss of life are no place to be playing around with community developed applications. If the space shuttle's computers crashed everytime my Linux box has (about once a year or so.. depending) it would be a national emergency. These systems must NOT fail. The code is tightly audited, new procedures have to be extremely well documented, etc. Now, the payment system for the New York metro is one thing.. but the air traffic control system is quite another! I do not want a bunch of 15 year old kids contributing code that will decide whether the plane I'm flying on will fall out of the sky or not. These systems should be redundant, expensive, and extremely tightly audited.
And like any good Windows NT implementation, they do crash. I've seen them bluescreen. If you open up the panel, you'll see that the outer frame of the machine covers the taskbar, with start menu at all.
Someone changed the default backdrop on one of the machines in a local station once. :)
At least now I know who those peons work for when they're in the station slowly servicing one of the malfunctioning machines. 'Death to Cubic!!' I can say, because I pay taxes. Damnit.
--
Do you, incidentally, know the current status of air traffic control computers? The system in use now runs only on hardware that doesn't exist anywhere else anymore and can't be replaced. There have been several attempts to replace it with a newer system, but none have worked. It's a reasonably difficult system to build the basic functionality for, and then there are a ton of features that the users would like, but the system can never crash, and has to be able to deal with various disasters without crippling anything.
It's precisely the sort of project that a company will tend to get bogged down on writing and then be unable to debug.
I don't think it's a terribly good candidate for an open-source project, though, since open-source is generally good when there are programmers who actually want a copy of the program. On the other hand, I don't see any reason to have it be closed-source. It's not like the company is going to have to deal with competitors stealing their code or have to track down unlicensed installations.
The DoD is falling for Redmond in a big way. No way the DoD can obtain the right to improve on what they get or let other parties improve on it. Just do Microsoft's bidding like everybody else.
-cjr
Not necessarily. For something like the Metrocard system the necessity of open source is not clear to me. The government would be much better off mandating the use of hardware available from multiple vendors and owning all of the rights to the software. That makes sense. If one vendor goes belly up or tries to negotiate for a lot more money when its renewal time, the government can go to another vendor with the software and ask them to continue support.
Open source? No. Owned source? Sure.
One of the biggest benefits of open source software is that its cost is infinitely scalable. How much does it cost to install debian on a compuer? Nothing. Copy the disk from a friend and install it for free and completely legally. How much does it cost to install debian on 1,000 computers? Nothing. Still free. Use the same disk if you want to. Now run the same test for a Microsoft solution. The Linux solution is infinitely less expensive.
For corporations who have to deal with competition this is an important factor. But for government, cost is hardly a factor at all. Government is not competing with anyone. If government does not have enough money, then it just takes more money from the taxpayers. Anyone here who has worked in a government job can attest to the enormous amounts of waste becuase of this and other aspects of government work.
So in relation to the post to which I'm responding: sure the NT solution is infinitely more expensive, but this is a government job. It's not like the government is going out of business due to any competition. Taking that pesky little "cost" aspect out of the software decision makes the decision much easier: Go for the convenient solution and make the taxpayers sholder the cost.
Vote Libertarian!
I don't make the rules. I just make fun of them.
There's also a complete hold on Windows 2000 deployment across the DoD right now until the impact of migration can be understood.
So CVN- 77 is on hold? Or is the Win2k testing time requirement already factored into the shipbuilding timetable?
Your Working Boy,
---
Just because 3 high schools in your area didn't immediately get excited doesn't mean that nobody would be interested...
---
---
When NYC first started to offer those 7 day and 30 day unlimited metrocards the windows NT based system was crapping out constantly because of all the new transactions the system was doing. Wonder how a linux or bsd based system would have handled it....
***There is no point in asking, you'll get no reply***
You are correct in that the DoD does not require all code to be made available to them. However, a large percentage (I would say a majority) of custom-code contracts require that the DoD have access to the source code. In my experience with the DoD, particularly in the security field, it is mandatory for certain segments (the NSA would like to inspect your code, please...). In other parts, code access is only allowed if the company goes out of business or stops supporting the product (where "supporting" has well-defined limits). In stil others (particularly when buying mass-market software), there is no requirement for code access.
I know I'm being anal-retentive about this reply, but then again, so are you.
;-)
-Erik
There are always four sides to every story: your side, their side, the truth, and what really happened.
Apparently, someone else out there agrees with you :-) In fact, when I mod I generally do exactly what the poster asks me to do: at least when they mark their own posts flame bait or whatnot. I certainly didn't expect my little cold-medecine induced rant to get modded up to +4(Insightful). I think those moderators were on more drugs than I am ;-)
On the subject though, using the GPL/BSD/Random Open Source license is completely not necessary to prevent vendor-lock in. The solution is simple: contract that the airport will own the code and they can do whatever they damn well please with it, wrt to having someone else work on it. I was mainly ranting about the fact that a "sourceforge" project, started and run by hobbiest who *haven't* studied what is necessary in a control program, had better not get anywhere close the my airport. If some company wants to release the source code to a program already in use to expose bugs, that's cool, but that isn't what Michael is shooting for. Well, I gotta go, but I enjoyed your post.
SCO always turns up in the most unusual places, it seems to give the warm fuzzies to large embedded systems people. I've seen it in telephone systems, satellite equipment, automobile diagnostics, etc.
-- Remember: Wherever you go, there you are!
actually, what would be the benefit of opensource... not inventing the wheel a hundred times a year ! I'm sure if it was open source all airports in the world could benefit from it. The same especially goes for the metrocard system. I mean if one good system is on the table, why not use it everywhere ?
New things are always on the horizon
Just because something is open-source, doesn't mean that the Gov't *has* to use patches from J. Random Coder in its "official" tree. There's a spectrum of open source-ish licenses, but they have (at least) two important aspects in this discussion.
Disclosure of source:
You're taxes are paying for this stuff, remember, so you [should] have a right to see how the entire system works. There are several examples of disclosed-source proprietary licenses, which might be a middle ground for a government-mandated source disclosure programme and a rabidly anti-sharing anti-caring american-typical oligopolistic corporation.
Provision for forking/patching of source:
Your taxes have paid for this stuff, so you should be able to use the source. If you REALLY want to, you should be free to start your own tree based on the theirs, or, when you are using the code, apply you own patches. At the same time, that does not mean that the Gov't should necessarily incorporate every patch into its own source tree. e.g. Linus doesn't have to accept Linux patches from you, but, thanks to the GPL, he can't stop you taking a copy of his source tree and patching it, and releasing it publically. Now think of some Gov't department. They don't have to use a patch you think would be a good idea for their software, in *their* build tree - but if they also think it's a good idea, they can fold it into their tree. So, software used by the gov't could be open source, but they would, perhaps, prefer a cathedral-ish approach (e.g. gcc, rather than a bazaar model - e.g. all the little applets that make up gnome-* or kde-*). Thus, I'd say open-source would be ESPECIALLY good for critcal systems, since there'd be the benefit of what is effectively massive peer review. The motivation would be there for third party contributors, who have a definite interest in a smoothly running system.
Choice of masters is not freedom.
Specialisation usually leads to beneficial economic effects. What is the failure in this case is the lack of managers who can pay attention to detail and in specifying the contractal obligations of the ssytem (a more extreme view is that governments by definition are inefficient at allocating resources compared with private markets due to information asymmetry). One can argue that a plane cannot replace a boat but then when everything became standardised into TEUs (Trailer equivalent units) people were able to compare the time/cost tradeoffs and make calculated decisions. Unfortunately we haven't yet reached that stage in software engineering where one can look at a piece of code and determine its efficiency.
LL
No, it's capitalism. The company *sells* its work on the software to you, instead of "renting" it like software licensing normal works. Don't be a twit.
And the code review process would look like what exactly?
Indeed. And Open Source doesn't have to be "free beer" if you don't want it to be. Suppose the US Department of Transportation (DoT) were to put out a contract for a new ATCS on the condition that it be Open Source compliant. The DoT might pay tens of millions of dollars for such a system. Regional airport authorities would have the ability to make (or contract out) bug fixes or enhancements as needed, although they'd probably prefer to coordinate through the DoT.
This is a situation where you don't care if it's "free beer" as long as you can have "free speech."
The main reasons for making something closed-source is so you can stop people from hacking it or giving copies away.
Government has no need for this. I can't think of any good reason for making something closed-source in the first place.
Now licensing something under, say, the BSD license, isn't the same as having an open development model. For some software, it would be so specialised that the resources spent on setting up websites, mailing lists, CVS servers, etc. just wouldn't be worth it. But others would be free to do that if the software was generally useful outside of the government, and the license was right.
Seems that math, chemistry, & physics would especially be well suited to computer based instruction. Anything requiring visualization to perform the calculation could be enhanced by viewing multiple animations right next to the math (simple algebra and vectors...maybe some calculus).
The only answer I can think of would be cost, as a company is likely to charge more to do the development if they have to give up control of their source code. That should simply be factored into the bid process, with the notion that a little extra money spent now will be an investment and not an expense. Government is vast, so I don't see any project being a one-off in isolation. Card readers, for example, and subway card readers in particular, aren't the problem of one city, but multiple cities around the country. They would benefit if they could all base their implementation off some common base. It's such a reasonable idea, you just know it'll never be done . . .
Note that the DoD did require the GPL when they funded the development of the GNU Ada front-end to GCC, aka GNAT, a while ago (circa 1995).
Now GNAT is of course still free software and ACT and ACT Europe are continuing the development and offer paying support services. Up to now they released sources from time to time, but it looks like the Ada front-end sources will be included soon in the CVS GCC repository, see the discussions on the GCC mailing list.
Disclaimer: I worked for ACT, but I no longer do, and at work I'm a client of ACT support services.
I've seen a lot of people post about how it would be a risk for mission critical operations, but what about the general risk of making the source readable to all. It should not be overlooked that some random person with a grudge against the US Government might get his/her hands on the source, and it is likely that any flaws that they find could be used destructively against the project before the bug is found by the developer community. It could be quite dangerous if some script kiddie in decided to totally screw over the FAA's software.
The DOD doesn't require that. That information is wrong. For instance an application used to order parts for the same DOD you talk of, the corp retains FULL CONTROL of it in all aspects. All we do is pay for it with tax payers money. Every year. This said app is one of the major things that fuels the army and I myself in a couple of months could probably write something better.
Yesterday I was coming from work at the Spring St Station In Manhattan(SOHO). It was my first week at a new job and that first friday always seems so HOORAYISH! In any event I pull out my metrocard which is unlimited and swipe thinking nothing of it. I push the big obtrusive turnstyle (granted they don't have any booth operators there) and it stops. So I don't get through. I try again and it says that my card has been already used which means I must wait 20 minutes to get through again AND/OR spend another 1.50 to get through.
Basically all this means is that their are problems with the system AND yes the MTA specifically would do best in probably using an opensource system as they are held by the belt by a corporate entity.
The thing that makes me wonder though is that these corporate entities have enough power and cash reserves to train their own in-house people to CREATE their own system. This way they spend alot less spending through the nose, They get qualified people who are happy and their own system and everyone wins. This also applys to many other corp/gov entities. Take for instance the army; they outsource most of their technical computer work!! How stupid is that?! They take all that time training people in their MOS (Military Occupational Skill) and they can't just makeup a new one MOS for computer programmers or tech guys to write their own apps?! It's stupid. In any event.. Enough ranting. If you corps wanna pay through the nose, and be held by the balls by some other corp; so be it. I'll just be sure not to invest in you as my money isn't being put to good use.
That's what digital signatures are for. It wouldn't be difficult to make a system that would digitally sign each card, and only accept cards bearing its own signature. The source-code for the system would still be open-source; the only secret would be the system's private signature key.
And while I'm at it, the NYC metrocard system is poorly designed from a mechanical point of view. It takes a considerable amount of practice to get the cards to swipe properly at the turnstile. Old folks seem to have quite a lot of trouble with it. Obviously this is because it's hard to make a system that works well with cards swiped at variable speeds. Why didn't they go with a system that sucks the card in one slot and spits it out another slot, like in London and Paris? Whatever one can say about the deficiencies of the latter two transport systems, their turnstiles work every time.
Ben
--
If a portion of the code is a trade secret, hard code those specific programs to a plug-in board.
All the rest of the code is open. Your hardware doesn't have to be.
Later
Erik Z
Democrats or Republicans. They are both taking us to the same place and they are not afraid of us anymore.
Why not start with a middle ground? Simply stipulate that the contractors must provide source code with all deliverables, and include a perpetual non-exclusive right for the government agency to use and maintain the software.
A dingo ate my sig...
Let's see, currently government software is written by paid professionals and audited professionals. (at least where it counts, ala FAA).
If the process were open source, they would still have to pay people to write it and pay people to check it. You think that volunteers are going to audit air traffic control systems, or that I would trust them to?
This is sort of like an intellectual property type thing. If you have only one client (eg, the FAA), you need some sort of control over the future of your specific contract. Otherwise, you get ready to release v2.0 and your competitor has stolen your code and stolen your contract. The competitor essentially can take the capital that I spent on the first version and use it to develop a 2.0.
This is a case where the government agrees to issue a monopoly over a specific market (ie, development of a contract). Companies want to be profitable, and locking down a few years for a government agency is a good start. Without the monopolies, the cost would have to be higher to entice anyone to undertake the project.
Troll Like a Champion Today
...even the NYC subway gadflies concede that Metrocard works and has brought benefits to the average bus and subway passenger. Obviously there are going to be problems with a system the size of the MTA, but there is certainly no great outcry against it or statistical case stating that it needs to be rebuilt from the tech level up. I'm as open-source as the next guy (maybe more), but I don't think there's an organization using solely open source technology that's ready to win and execute a contract this size today. Maybe someday, maybe someday soon, but we're talking about a massive engineering job here, and finding a contractor with the ability to manage that huge a process to schedule and budget is exactly the sort of thing you don't want to get caught pinching pennies on if you're a Transit Department desk jockey. The Village Voice article mentions no organizations who claim the ability to do a better job, much less with open source tech. The only alternative mentioned is union labor... and whatever one's feelings about unionized labor's real or idealized relationship to technology culture, the union representative quoted sure doesn't sound too knowledgable. Does anyone think he might just want the headcount back under his control, regardless of the results?
On strictly Metrocard tech terms, if someone wants to break the security on the cards, I might be more inclined to beleive that there are fundamental reasons to rethink the underlying technology. I've spent a lot of time working over the cards myself and with friends, and despite having a reasonable ear to the ground here, I have not heard of a usable, repeatable hack on them for years. The only hack I have ever verified myself was within the first year of the system's deployment and could get you one free ride, after which the card was rendered unusable. That hole was closed years ago. If anyone has actual evidence to the contrary, I'd love to hear it.
As regards all the points about how silly government applications of technology can be, well, duh. The article cited is about a specefic case, though, and despite my best efforts and poetic sense of justice, the thing in question works and at a reasonable cost. Anyone who thinks $390 million is a lot of money for the job and thinks that the city is being rooked should consider that at 3 million riders a day at a dollar fifty per trip, Cubic's tab comes to 85 days of revenue from a system that's been up for nine plus years. I'm not sure that on Planet Earth in the year 2000 you could do too much better.
In short, the article really seems to be looking for a problem where, if one exists, it's pretty far under the noise threshold.
Let me begin by stating that anyone that understands anything at all about US Air Traffic Control (ATC) knows one thing: the system is enormously vast, varied, and complex. And that would be true if one were just considering the types of equipment used, and the way in which they're interconnected. Include the demands and pressures placed upon the Agency that operates it, and a few more layers of convolution are added. The argument for throwing into that mix a demand for "Open Source" principles would have to be extemely compelling. And, of course, it's usually impossible to make a compelling argument without understanding the facts, which brings us back to the complexity of the ATC system, and the scarcity of those intimately familar with it. In other words, it's easy for someone that knows nothing of the system to suggest all sorts of solutions, and I'm constantly reminded of that fact nearly every time I read an article written by an "Aviation-Industry Journalist" and occasionally while reading posts here on /.
Here's a single, small, and simplified example of the issues involved:
The FAA is no in the process of purchasing over 100 new radar systems, known as the ASR-11, from Raytheon, a company that's been a leader in the Radar business for 50 years. The Department of Defense is purchasing approximately 100 more. Each system has, at it's core, two Sun Sparc 5 workstations (soon to be upgraded to UltraSparc). Of course, the workstations run on Solaris. However, the software that controls the radar and performs much of the data processing, is a proprietary product of Raytheon. Although it's likely that the FAA will gain the rights to the source code, they'll (you'll) pay dearly for it. What's more, some portions of the code, perhaps not in that software, but in other parts of the radar, are essentially "trade secrets" and Raytheon would be loathe to part with it at any price, particularly if they knew that it would be released to the public and, of course, their competitors.
How might Open-Source principles be implemented in this case? Raytheon engineers are the experts, and the FAA wisely allows them to design the hardware and write the code. Are there those that would suggest that all of the Radar's code should be made available to the public once purchased, at great cost, from Raytheon? Or is it a matter of writing a contract that demands that the Manufacturer use open source principles in every aspect of Radar design, and/or that the radar itself use open source applications to perform ever function? Would some suggest that, when planning the purchase of future Radar systems, that Raytheon design the hardware and the FAA write the code?
Finally, and perhaps most significantly, in this case how would Open-Source principles benefit the FAA and the public?
Who will spend time writing this stuff? The airlines. They lose tens of millions of dollars because we're using outdated technology for ATC. If they can improve the systems themselves, adding the features that they want, you might find things like flight corridors going away. Right now, these changes have to be lobbied for politically and pressure can only be applied indirectly.
It's a classic open source itch scratching situation, only its firms this time, not individual hackers for the most part. Though I do expect that individual hackers caught in their 5th hour of unexplained delays might decide to improve/write an interface to improve passenger reporting of problems via the ATC system.
In other words, qualified people are likely to do it, jumpstarting a process of upgrading the ATC systems in the US that has been horribly blown to date.
DB
You truly like inconsistent interfaces and consider them good UI? I doubt it troll.
DB
Oh, let's see (can't resist a few pet peeves), a proper universal clipboard where all apps can cut, copy, & paste so that terminal windows can copy, not only paste, the ability to have it always be ctrl-x for cut or alt-x for cut, not one way in one program and another way in another program, you know, little basic useability things that those closed source Apple folks have had down pat since the early 80's.
Can't we all just get along?
DB
The fair measurement isn't whether the stability/longevity advantages of open source hold true for "every conceivable software project". Rather it is whether the advantage holds true in such a large number of cases that it saves time and money to make open source the default.
I suspect that there will always be closed source in government software, specifically in military and intelligence applications where we certainly want our enemies to do their own darn R&D (fill in your own country here, you anti-USA-centric whiners). There may be other specialized cases as well (would we want wiretap software to be open source?). But those examples are irrelevant to the vast majority of computer applications in government work which would do just fine in open source.
DB
Actually, it is quite likely that the airlines would be major contributors. If throwing a thousand programmers for a year at the ATC system increases the number of planes that can fly out of JFK, LAX, or any other overcrowded airport by 5%, it's a net positive to their bottom line in less than a year. When planes sit on runways waiting for a slot or even worse, are kept in holding patterns, it's a very ugly cash hemorrage.
DB
Just because today, government specs are all incompatible doesn't mean that they have to stay that way and OS could be a method of achieving a bit more sanity. New municipalities are created every year. If even a significant minority were to adhere to a generalized OS software philosophy, it could reduce costs significantly. Existing governments would come on board over time (the whole process would likely take decades). My point is that if you don't start, you will never finish.
DB
ok i'm not a big linux geek so i don't know but i thought this was an x thing not an e thing? am i wrong?
Open sourcing government projects is a good idea, but like many ideas which make good (common) sense, bureaucrats by and large do not think in the same way we do. For them, it is not so much a matter of cost, but a matter of responsibility and who's to blame if something goes wrong. They live in a world where success does not bring anything (barring more work) and mistakes can cost their jobs.
In other words, given the choice of spending one million but shouldering any potential blame and spending ten million and out-sourcing it, any self preserving bureaucrat would argue for the latter, as if something goes wrong, they can blame the contractor. Whereas in the first instance, they would end up on the wrong side of a disciplinary committee hearing.
Karma makes sense. It makes a lot more sense if you add reincarnation.
It's simple develop packaged keybinfings and apply them everywhere.
HMMM
KBML.
Gah whiners.
The message on the other side of this sig is false.
So create a better open source business bureau.
I think we call those mailing lists, maybe you've heard of them.
I can say that linux-kernel list contains the most hotly debated content and commentary I've ever seen.
The thing generates several megs of text a week.
Judging that an 80x25 xterm contains a maximum of 400 characters and that most messages on the list fit in about ten of them, 1 meg = 250 messages. That doesn't usually include code only links to it as the message is mostly an over view and the discussion is in the code comments.
The message on the other side of this sig is false.
Problems:
You honestly don't want people to be able to check the quality of the traffic control software?
If we have two apples both copied from the same genes.and I bite into the one that is in my hand, are you saying I just infected yours with my microbes?
As someone else said you don't have to accept patches.
Linus manages his work like a cathedral but allows submissions to be made like in a bazaar.
The message on the other side of this sig is false.
..was mainly ranting about the fact that a "sourceforge" project, started and run by hobbiest who *haven't* studied what is necessary in a control program,...
What in the fucking world makes you think open source means an end to review?
You must think no software gets reviewed such as the alternative. Otherwise I have to accuse of being inexplicably biased.
It's like some black cloud forms when you see the words open source and lightning strikes the first thought that comes to mind.
I think the cough medicine has had a serious effect on your logic.
If some company wants to release the source code to a program already in use to expose bugs,
What the fuck is the difference between releasing source code and releasing source code to used programs. Code is code. Bugs are bugs.
Geez next you'll want a cordless extension cord.
The message on the other side of this sig is false.
Whose right is to stop people from entering markets?
How about adapting...
The message on the other side of this sig is false.
That's not obscure, it's secure. Only you know the system used to encode that data (although, I suspect it's a direct substitution type code, where something is simply replaced by something else according to a predefined look-up table.) Now, if this look-up table were published somewhere, or available for the asking if you could figure out who to ask, and how to ask them, then yes... that would be Security through Obscurity - The belief that noone who doesn't deserve to have access to the data would go to the trouble of working out how to get it. Otherwise it's just encryption. (An strange sort of encryption, to be sure, but still encryption)
NipokNek
Why choose white shoes?
Well, since you decided to post as an AC, no one cares what you love.
Oh, and also... Open Source projects are NEVER done. When will Linux be done? The whole point of not closing the code is that the code never closes. If it did, it would just be Closed Source programming with a lot of Beta testers. If you don't even understand THAT much about it, just cut your keyboard cable now.
Nipok_Nek
Why choose white shoes?
Errr, software projects DO get done, be they open or closed.
Would anyone care to argue that Windows for Workgroups 3.11 is NOT a done project?
How about FreeBSD 2.2.8?
Redhat 6.1?
All of these projects are done. Some of the code and the ideas have moved onto the newest version, and is now shipping in the forms of Windows 2000/WinME, FreeBSD 4.1.1, and RedHat 7.0
Projects are DONE when the functionality wanted is achieved, or the people doing the project stop on the project. It may not be done correctly at that time, or work as one might want, but that is a different can of cockroaches.
If it was said on slashdot, it MUST be true!
Today's vices may be tomorrow's virtues.
Okay, I understand the reliability of Linux is very high but that's not true for all open source projects. Anytime you have a large pool of programmers working on code you're going to have a large difference in programmer skill levels.
Isn't it likely that bugs are going to slip into systems that can't have them? Okay, the system already has bugs but at least they have somebody to back them up and fix them. Yes these arguments have been used against open source before but this time we're dealing with large and boring systems.
Who's going to spend their free time writing this stuff? Wouldn't it be tempting to walk away as soon as you realize that it's not fun? You have to pay people to work one large boring systems.
Vanguard
That which does not kill me only makes me whinier
Pilot: This is flight CN-123 flying Beijing-New York. Do you copy?
Control: Uhm... Yeah... Hang on... To himself. Why aren't you on my screen?.. Wait just a sec... (typing in "deja.com" in his browser and searching for "China airplane linux"). Oh, I see... Multibyte support is lacking in the version we're using. To the pilot: OK, can you do some boxes while I download the latest CVS?
If you open yourself to the foo, You and foo become one.
Isnt equating using powerpoint presentations to get your jobs done an oxymoron or something?
When FOIA request results are released, they blank some stuff out. I can only assume that is what he refers to in encryption.
Pax Digitalia
-= Griffis =-
Try http://www.rounin.com
Al Gore seems to think so.
| Ceci n'est pas une pipe.
Another way to look at this is: the government is throwing our money away on high priced, low quality (98,NT) Software.
As taxpayers, we should demand from our representatives that this waste be stopped immediatly. This money could be better used for Social Security or the Federal Deficit for example.
Paying millions upon millions of dollars for software when there are high quality, free alternatives (Linux, *BSD, etc.) is irresponsable at best.
Why do so many people rip on Mozilla? I think it's great.
-=) A Good Kingdom is Coming (=-
No way! Surely government contracts like this should specify open source to avoid technology lock-in. We need a transition in mentality, but I hardly see air traffic control systems as something well suited for experiments like this. Nor do I see that it needs to be changed unless it fails. Let's just keep air traffic safe and sound for now, shall we...
Unfortunately, graft is the American way. I'm sure the government is one of Microsoft's biggest customers, if not the absolute biggest, and Microsoft can certainly afford to buy clout in Washinton.
It's really unfortunate, because all that money being spent on Word, vs. saved by something like Staroffice, could pay a lot more teachers, build highways, etc.
Also, the bickering over who gets to be the supplier wastes time, and prevents vital things from being done. I can think of one case where a system for putting altitude transponders in aircraft was stalled for over 10 years, so certain politicians' "patrons" could come up with their own, and get the contract. In the meantime, scores of lives were lost in mid air collisions that could have easily been avoided.
this is not socialistic in any way, shape, or form. social security is socalistic. medicare is socialistic. this is simply letting the work be done by the lowest bidder(s) (in this case the open source community) a group who has proven their ability to code circles around most busnesses. socialisum is "redistributing wealth" i dont see any wealth being redistributed... does anyone else?
Why shouldn't government software be open-source? We pay for the government's very existence through tax funds. We should be able to see what the developers are doing and contribute to that. It's a way for (1) the government to escape the stranglehold of money-hungry private consultants (note that I said money-hungry, if you are not money hungry, you are exempted from that statement) like the MetroCard fiasco and (2) developers can do something to help their commmunity. Everybody wins...except the aforementioned money-hungry consultants.
the deacon...that's all you need to know for now
Yes, the government should go open source. For everything that isn't life-threateningly-important that is. Subway systems? No. Air traffic control? No. School software? Yes. HSFIS (head start child management software), yes. 95% of software paid for by the government. Yes. That other 5%? Well, I'd prefer to leave the DOD code alone, and other sensative sectors.
Daniel
At least with Federal (U.S.) Goverment tax-payer paid for projects, a Freedom Of Information Act request is all that's required to get the source to custom-developed-for-the-Fed software. And FOIA requests aren't all that hard to file.
---
Guillaume
give me all your garmonbozia
I like the X selection mechanism.
TO BUY A NEW CAR WOULD MAKE YOU SEXUALLY ATTRACTIVE.
What makes this so special because it involves computers?
"How's this. Government choses a contractor to build a system. Contractor delivers code to government, government releases code as open source. Government then can choose from any interested parties to manage the code base for them, do one-off mods to the system, etc."
While the popular term, "security through obscurity" isn't actually a *good* thing, consider this. As the government, or a government sector, would you want any security flaw in your software to be openly available to the general public knowing they, and only they, will be the first ones to find that flaw? Atleast with contract developement, your security problems (and there are *always* security problems) are relatively much much harder to find, and there is always the possability they will be realized before the general public finds them, a la contracted hackers or the original developement company.
Ace
Open source works because the community of testers and debuggers is so large. It won't work for ATC, simply because there are so few users, particularly computer-sophisticated users. What may work is the use of open-source components in air traffic control applications. The software reliability needed in ATC (about 2 hours of unplanned down-time per year per system) is comparable with that seen in some open source products and several times what MS products seem to provide.
Air traffic controller software probably IS open source. I know for a fact that the algorithms for TCAS (Traffic Alert and Collision Avoidance System) developed by MITRE are open source, although the implementations by the manufacturers of the cockpit devices may be proprietary.
I worked making software by contract for various PA gov't agencies, our contract stipulated that our software was public domain and we even had to send on our source for archive. What's missing: a centralized store of gov't software to reduce duplication of efforts -- an interesting project, if nothing else.
As opposed to the closed source world, where companies go out of buisiness, stable versions are replaced with less stable versions, and prices can increase.
The result is that the more people who need a product, and the more they need it to be stable over time, the more it makes sense to open source it. The greatest open source projects are the most ubiquitous. Gui cd players are all well and good, but 'less' is universal, and without competition.
Open source friend is a programmer's best friend. We all know the benefits it has had in the industry. The only problem with it is that is it not very easy to make money off of. The air-traffic control system is an incredibly complicated system. Although it may be easy to make it open, one of the ways that money is regained is by selling (much like most government projects). In fact, the US just cut a deal with Norway for some rediculas amount to sell part of the Defense System. If it were open source, then it could not be resold. I don't think there are too many good ways to make that much money from open source code so until that day, don't expect government projects to switch over.
int func(int a);
func((b += 3, b));
It needs to hire people that know what they're doing. If any part of the government is going to be involved with customized software or hardware, there should be people in-house to handle those needs. Maybe not at the local level, but at least at the state a federal levels. The fact that jobs get contracted out to the lowest bidder is stupid, as many things in government are.
/. to condemn the government for having inadequate security, since we deal with it on a regular basis. Don't get me wrong, I'm not standing up for them, but websites shouldn't get the governments primary responsibility. Everybody's getting caught up in the web, and they're trying to turn it into something magical instead of something functional. Look at Samsung or HP. They chose to focus on making a static image fairly appealing. However, it's not functional.
One of the contributing problems is that nobody wants to work for the government due to the red tape. Those that do, only do so mainly because they see the money, and realize that there is nobody there to verify their work.
Example: Local town clerks office hired a consultant group to network all the machines together and setup a way to get e-mail through just one computer on a dial-up. The consultants installed Exchange server and then couldn't figure out how to set it up. So, they called the company I work for (ISP that provided the dial-up) and expected us to spend hours on the phone trying to get Exchange server configured. The consultants were on that job for over 4 weeks, and I don't know the exact cost, but I would assume at least $35+ an hour.
Anybody that has "Consultant" in their title should automatically be disqualified from working on any technological project.
Back to the topic though. What would be the purpose of the government open-sourcing all it's code. It's not "their" code. The code was created by somebody that was awarded a contract to do the job. Who would manage the code if it was opened. If the government appointed somebody to manage the code, what are the chances that they would be qualified enough. They could let some hackers back-door into the code base and where would that get us? Then we'd need a peer review system, and who would be part of that?
Truth is, the government should spend more resources on doing whatever needs to be done to make everything run more effeciently throughout the country. It's easy for everybody on
Ok, it's late and I'm out of caffiene. I think I spelled out enough problems here, and while they probably aren't grammatically correct, or show a proper thought process, I hope you get what I'm saying. Basically, the whole system sucks, and who freaking cares, because whatever they do, we complain about, and whatever we do, they complain about.
ok i'm not a big linux geek so i don't know but i thought this was an x thing not an e thing? am i wrong?
:P
You are correct, the original poster is just ranting unintelligently.
The X windowing system is pretty poo, but E and Gnome make very good use of it and make it quite usable
Anyway I'll shut up, this is too offtopic..
What exactly about Software For The People, By The People is socialist or communist ?
That's what happens out in the commercial world. When we outsource, and the code is bespoke, then we, the customer, own the code and retain the right to get a different company to maintain it if the original company screws up.
If the outsourced code is a package, we demand the code be escrowed. If the company fails to deliver on it's contractual obligations, the code gets released to us to use to continue our business, but not for any other purposes.
If you're big enough, some vendors even give you the code to play with (e.g. IBM), for an extra fee, and under restricted license conditions.
If the bidder refuses to play by our rules, they don't get to play. After all, we hold the dollars :-)
It's not really an open source versus closed source argument, it's really about a very poorly negotiated contract, sadly a far too common occurrence when it comes to governments around the world.
Having stood on both sides of the out-source relationship, I can state that maintenance costs are seldom if ever cheaper if you out-source :-) That is, IF you already have an IT shop that is competently run.
It is certainly cheaper if you include the cost of lost opportunity in your given market. The key factor for the bank I worked at in deciding to pull our systems back in-house was the huge time delays we faced in getting changes made to our out-sourced systems. The delays seriously affected our ability to respond to changes in a volitile marketplace and was costing us market share. Now, the only things they out-source are the commodity parts of the systems, the parts that never change. Anything else stays firmly in-house where people who know what they are doing look after them.
There are two areas where out-sourcing does make economic sense. (1) If your system rarely changes, and all you need is a box to run it on and day to day support staff, out-sourcing is un-beatable. (2) If you don't have an IT shop, building a good one costs a lot of money and time, and needs a lot of experience. Getting an out-sourcer to build and run it for you for you makes sense until you are mature enough to take things over.
How's this. Government choses a contractor to build a system. Contractor delivers code to government, government releases code as open source. Government then can choose from any interested parties to manage the code base for them, do one-off mods to the system, etc. The key here is that the code is open for all to see, so you can hold competitive bidding for all work on the system. Company X wants to get a slice of the governments money? Download the code, assign a few programmers, and start bidding. The work is still done by reputable firms, to exacting standards, and carefully reviewed before being checked in, but the government isn't locked into a single vendor.
Think of it like building a house. Just because Builder Bob did the kitchen, doesn't mean he gets to do any re-modelling work in 5 years time. I choose who I think will do the job best, and that probably won't be DIY Joe from next-door.
OK, so it's not pure Open Source as we know it, and I'm sure people will pick holes, but it's a high-level concept, and it sure beats the current way of doing things.
And it sure sounds like how Linus runs the kernal...
But I thought Governor Bush's "educational reform" in Texas fixed all that ;)
I think it would be beneficial if students could log in over the Web and check their current grades, so that grades on their report cards don't come as a surprise.
I think it would be beneficial if students could log in over the Web and change their current grades, so that grades on their report cards don't come as a surprise.
I'll see your senator, and I'll raise you two judges.
ps. I suspect you're hitting the middle row of your keyboard, and it's not a code at all. You lied to me anonymous coward, and i'll never trust you again.
--Giving to trolls for the benefit of us all
> It will keep attendance and tardies,
They let you keep track of 'tards? That kicks ass!
--Giving to trolls for the benefit of us all
Reading all this stuff makes me giggle somewhat, I spent many years driving a truck, the idea of my boss not wanting me to know how my truck worked, gives me belly giggles.
Have a look at.
http://www.dollyknot.com
How do I get it to appear as a clickable connection? Typical.
The secret is, who has configuration power?
Just flying an idea, like a kite. Why not explore the hardware solution to the security problem, make the processor identify the human controller via retina patterns, smell, fingerprints etc. the more accurate the process of identity, the more the security. In other words, stand up and be counted.
Pete.
It's called an elephant's trunk whereas it is in fact, an elephant's nose, a nose by any other name would smell as sweet
The day that happens is the day I'll stop flying.
"Air traffic control systems upgraded to RedHat 7.0" - three weeks later: "planes drop from sky".
Open source is nice. It is NOT the solution to everything.
You may be surprised to discover that Eric Raymond agrees with you.
The Magic Cauldron: When To Be Open, When To Be Closed
-- Nick "Hallo this is Beel Gates, und I pronounce weendows as
I think that they should spend millions of dollars to stay one step ahead, not spend millions of dollars to stay one step behind (as it is now). If this were the case, Carnivore would've been implemented four years ago. Let's face it, the government is basically a society of Luddites. Sure, there's lots of hype about what new technologies they're moving towards, but look at them now. Government websites running off of NT 3.51. FBI offices with Olivetti stations at every desk. As it is now, there is no such thing as the Internet Cops; just as it was in the 20s and 30s, the criminals have the better vehicles and weapons. Hackers are equipped with the technological versions of V-8 speedsters and Thomas submachineguns. It's time for the government to form a strike force that will change Internet crime in the same way the FBI changed physical crime.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
but even still, if the goverment did go to OSS, it would result in much more efficiency at a much lower price...
i'd be very willing to bet that if the gov't petitioned the OSS community to help write a system (assuming nonintrusive like the Carnivore) there would be MANY willing contributors, who would possibly also write it for free (similar to the Mozilla project)
--------------
I think the most common concern about releasing security sensitive source code (like metrocard, or anything related to transactions) is... "Hey, it'll be easier to crack if they can see how it works". It's a bit of a hurdle to overcome. Especially if we start talking about banking systems. The problem is no system can ever be completely secure. There will always be some way to break it, due to the nature of complex systems. The reasoning behind larger, more secretive systems is that if there will always be some flaws, let's not give them any easy way to find them. In some cases, if the code is reasonably secure to begin with, or if the application is extremely sensitive, open-sourcing doesn't really make sense. If the company can turn around security fixes in hours or days, and not weeks and months, then it works pretty well.
So what you have is Social Engineering winning out over software loopholes. Amusing.
--
When you sympathize with stupidity, you start thinking like an idiot.
Open source has its places. Air traffic control is not one of them. With a highly critical systems, there needs to be some measure of accountiblity there.
-----------
end communication
Seems to me the lack of interest is the primary problem.
"// this is the most hacked, evil, bastardized thing I've ever seen. kjb"
Having been involved in contracting to the gummint for my entire professional life, let me tell you why government will never use open source to any great degree: Because there's no one to sell it to them ('sell' in the sense of being an advocate). For a corporation, there's a profit to be made, and the more custom the job, the bigger the profit. So "the government" is always going to have a horde of salesbeings hammering them on why they need a fancy-dancy custom solution, while the open source folks will have a few nerds writing the odd letter or two. Oh, sometimes you'll get an enlightened and energetic individual within government who will be the advocate, but it's rare. So generally speaking, there's no contest.
Why would information released under the FOIA need to be encrypted?
CEE5210S The signal SIGHUP was received.
I understand that portions of documents would be redacted, but surely they'd just remove those portions rather than provide encrypted copies of them to the public--I hope :). Thanks!
CEE5210S The signal SIGHUP was received.
int main() { int taxes; if (taxes > 10000) { cout "AUDIT"; forecloser(estate); } else if (taxes 10000) { cout "AUDIT"; auction(privateherlums); } return -1; }
prosebeforehos.com
Where Your Vote Should Go
great comedy company.
I think there's a point to be made that is the converse of the impossibility of overspecifing software. The reason governments spend $500 for a toilet seat is the same reason they will never accept an off-the-shelf solution, open-source or not. For instance, I distinctly remember the mil spec for an ashtray (of course called something silly and bureaucratic like "ash receptacle, glass") involving hitting the ashtray with a certain kind of hammer, on a certain kind of board, and counting the fragments to see if there were too many to meet the spec. Software that governments buy is the same way, put out for bid with a ludicrously detailed spec, because of course there is no room for common sense in the government procurement process. If the program meets the spec, the company gets paid, even if it has severe defects that are not explictly forbidden in the spec.
Now, maybe it would be to a government's advantage to require the source to accompany the binary, but certainly not for the purpose of sharing the product among different municipalities. They all have their own little twisted specs, which of course are all different and incompatible. Kind of reminds me of all the Open Source Licenses. Mostly do the same thing, but it doesn't stop them from being a huge pain the reconcile.
--Del
At my high school all the teachers use a program for Windows called WinSchool. It will keep attendance and tardies, average and print grades, and record and store them to a network hard drive. The downside is, I dont know the price to it. The teachers do seem to be able to use it just fine.
Microsoft is not the answer. Linux is the answer. Microsoft is the question.
This assumes the government wants to take direct control over certain services and or put projects under public control, which is essentially a socialistic concept. The US government generally likes to subcontract everything out to private companies, so these huge monolithic projects that run into the millions (billions in some cases) that also over their dead lines, are essentially just part of the system. Anything difference requires a change of ideology, and generally the American populace deplores more government control in their life's.
Nice idea, I'm sure many people on this post will make some sort of synergy to communism, it's not quite that extreme, but there's a lot of politics (and $$$) involved.
Overall the security would be increased. More eyes looking at the code means less bugs and less security holes. The code doesn't have to be public if it is open source, it just means that the MTA would have access to the source, and would be able to do in-house fixes, which would surely be cheaper, and more flexible. As for the negatives of this, they are all cancelled out by the pluses. Bad people who had access to code can more easily do harm, but then again good people can more easily fix it. As the encryption community has shown, closed systems prove poor security. All the transit systems shouldn't be forced to use once system, but they can have several competing systems, and in the end, the better systems will end up prevailing... Plus they can hire a bunch of poor computer science undergrads like me to mantain =)
who agrees whole hartedly with this article. Being an insider I know what it is like. Let me share with you the world of a county IT guy. First off, like all IT depts. we are under staffed. The county council is willing to spends hundred of thousands of dollars on a sinple project by outside vendors but are unwilling to hire in house staff that could do the same thing quiker and with less cost (my salary sucks).
Let me give you an example. My county recently partnered with a site to offer property tax payments online. Two county big wigs went to some conference where they were met by sales reps of this company. They were sold a bill of goods about how it could be implemented with no cost to the county. The company was invited to demo the product (I don't think IS staff were invited). The next the one of the county big wigs said we are going to do it (on a morning radio show). Now we are dumped with getting this project done quickly. Remember that this was not supposed to cost us anything but instead it took one or our programmers two months to get the complicated tax data in a format the site could use. Since it was introduced I think 4 people have paid their taxes with it. In a year when the contract is up they want to charge $12,000 a year. For that cost we could have developed it in house once.
This is just one of many examples where departments and county officials are sold on poducts by vendors with little or no cunsulting with the IS dept. Often times projects could be done in house (with a few more poorly paid staff members) using open source products and industry standards. Instead we are suck learning proprietary systems that don't work on our network and rarely work as sold.
We run a novell network, there have been a lot of products sold to the county that work only properly on NT networks. IS is supposed to approve IT purchases be we do so often under pressure from people above use.
To try and stop this I have proposed and developed a county standard to hook data is different systems together. It uses XML, HTTP, and TCP (all the proper buzzwords). Most department heads are on board but some still resist because their vendors will propably not want to implement it.
We are trying to use Linux more (DNS, SMTP ) and are continually having fights with MS licencing lawyers.
Bottom line, government IS staff often want open source but no one ever askes our opinion.
I think that's a great project. Look at all of the open source software out there that is so much more complicated. This stuff wouldn't be that bad, and the coders are right there as parents, and the husbands and wives of teachers. Consider, how few and lame the educational software is. Doesn't the sheer number of variants on the same theme scream for open source as the only answer? Great way to throw Open Source right at kids, too. They see it, they know Jimmy's dad made it, they know they can learn, too. Any educators out there want to lead up a project?
OSS isn't the issue here. NYC Transit has problems wihth the Metrocard because it bought into a custom system that it alone uses and that was developed by a company which (judging by the VV article) seems to be trying to equal Scientology for corrupt practices.
Transit should have bought an existing system, not one consisting solely of vapour. A good choice would be the one currently used in Japan -- although I admit to having no knowledge of its technical details, the system does seem to be uniform countrywide. It works well, is flexible enough for multisystem tickets, is fully automated (except for intercity reserved seats, which shouldn't be difficult to automate), and should be cheaper than implementing an entirely new system.
Of course, the US is infamous for NIH syndrome. It adopted GSM fairly late in the game (albeit earlier than Japan, which won't do so at all), and even then not as a countrywide universal standard. Then there's the Acela fiasco, where Bombardier was selected as the trainset supplier despite it having zero experience building anything capable of > 250 km/h operation -- and, according to their website, the top speed of the Acela trainset is only 240. Pathetic, to put it mildly. The current-generation trainsets used for ICE and Nozomi service have a maximum speed of 280 km/h.
At least we don't have John Howard or Jack Straw.
Nobody said it was anyone's right.
However, any significant reduction in the barrier-to-entry is generally a business consideration made when considering the value of opening internally developed and used software.
I never said this was a Good Thing, just that it's how decisions get made.
How is open-source air-traffic control software going to get tested?
The exact same way closed-source ATC software gets tested.
Just because the source is available doesn't mean that anyone outside the usual development team gets their changes accepted. What it DOES mean (the important part) is that if some other company can offer a better deal on the development while still keeping the gov't satisfied to the quality of their work, they can take the ATC software and (hopefully) the simulation environment used for testing and work on them themselves.
It's not like this is something Joe Programmer would be working on in SourceForge; rather, it's an effective means of preventing vendor lock-in.
Btw, I honestly think "(-1 Dumb)" or anything of the sort should get you a "(-1 Asked For It)".
Maintainance costs are seldom if ever cheaper when you go in-house.
Depends. If it'd be an entirely custom job and you already have folks in-house capable of doing it, I'd expect in-house to be cheaper.
Huh?
If SomeCorp(tm) contracts me to write a piece of software for them, and I do it, and they pay me, and we don't make it reeeally clear otherwise, it's a work for hire, and they own it, and I don't have any legal rights to it, and they can take the source and turn around and give it away or have someone else maintain it or whatever else.
That's the law. It's always been like that, open source or no. Unless your company agrees otherwise outright (and they usually do), they don't have any rights to the software anyhow.
I'm not sure that releasing ATC source will expose bugs. I mean, I wouldn't be reading it, not usually.
:)
Maybe if I were writing a game I might borrow a significant amount of exist code and debug it via that route... but that's about the only situation in which I, Random Joe Developer, would be messing with the thing. Remember that the article talks about funding these projects, open though they may be. It's not a hobbyist thing in this context.
Finally... even if the city owns the source and can bring it to a different vender, opening the source still helps in reducing lock-in -- if for no other reason than that it encourages alternate developers to get involved by giving them an easier way to look at the existing system; one of their developers can DL the code on a whim, without getting any suits involved, take a look at it and then take it upstairs. Furthermore, those relatively few developers who *do* look at the source for whatever reason (such as the aforementioned game example) benefit the sw's owner by increasing the number of people available for hire with knowledge of the codebase.
And finally... (with the exception of national-security-related sw) why the fsck not? With a company, there's a worry that releasing internal software will reduce competitors' cost to enter their market. I don't think anyone's going to worry too much about that when preparing to start a new government.
--
Maintainance costs are seldom if ever cheaper when you go in-house.
Obviously there is a lot of in-house developed and maintained software, and that's generally because it is specific to the way you do business and the excess cost can therefore be jstified.
I don't know where you came up with that baloney.
The Defense Department mandates that any software built for them has to also include the source code so that they can do whatever they want to it.
:)
They also mandate you write it in Ada...
Nobody suggested that "we" would be writing the code from scratch. The suggestion is that the existing code be opened up to our hundreds of thousands of eyeballs so that it can benefit from our feedback.
"The question of whether machines can think is no more interesting than [] whether submarines can swim" - Dijkstra
The arguement was since one agency is co-funding so many projects that it require the projects use a common software base and then the vendors build their hardware to support the software. The advantage would be no longer would NYC / LA / Boston / Chicago / SF / Etc. all keep reinventing the wheel nor paying vendors to reinvent (or resell) the wheel but rather one wheel be defined and used (with local variations) as needed.
The plusses are obvious: Municipal transit systems could cut their software costs. A larger common pool of code would distribute development & debugging costs. Hardware could become standardized. Bidding would become easier as systems would be more similar. Features developed for one municipality would become easily availiable to all, in return they would be reviewed & debugged, possibly being extended in turn.
Of course there are negatives too: Asking some some sort of board to come up with standards...
This could also lead to less-well built systems. While the core system may be standardized it'll have to interact with legacy systems for up to another 30 or 40 years in some cases. Those will all be unique or semi-unique ones and of course building bridges to them will have to be done locally.
Furthermore any common code base will have to be flexible enough to support the myriad local pecularities across the systems. It need not directly include support for these specific features but it'll require the hooks & robustness to accomodate them.
Finally this all reflects the competing free-market/governments-standards issues in the US. Presumably by letting various vendors freely compete they'll come out with better products faster & cheaper then any centrally organized government-run body could. That is official US doctrine.
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
Well, that's why I'd like to see some gov't funding for it. You aren't going to attract the Rastermans (Rastermen?) of the world to work on it - but it would be in the govt's best interest to pay people to develop it, and retain the rights to the source code...
---
"almost" arrested is like "almost" pregnant.
Michael, you didn't share some important details.
What really happened?
-fb Everything not expressly forbidden is now mandatory.
> there are about 25 people in our group capable of learning how to administer an NT box in a reasonable amount of time, there are about 4 or 5 available who have a reasonable background in *nix.
No, you had 25 people who could follow wizards, set file permissions, and maybe know how to invoke the registry editor to change a key mentioned in a book they come across. It's a start, but it doesnt make them qualified admins. NT administration is still tricky hard work that requires a lot of specialized knowledge no matter what the four-color glossy literature tells you. I find NT administration to be even *more* of a challenge at times, when I lack tools out of the box like truss to show me what files a program is opening (great way to notice things like "aha, it's looking at this old config file that should have been deleted"). I'm not saying it's not possible, I'm saying it still requires specialized knowledge and troubleshooting skills. You get those people out of enterprise environments, where you're *far* more likely to find an equal representation of unix admins as well.
I've finally had it: until slashdot gets article moderation, I am not coming back.
Apart from systems which are arguably for national defense, any software developped for or by the government is open source.
Its your tax dollars that were spent. The Freedom of Information Act here and other laws in other countries, clearly state that what's your's is yours That includes the software that controls the operation of any device.
If specific software was developped for specific devices that are bought by public funds, the source code to that software is also part of the purchase. That implies absolutely no warrantees on the part of the vendor. There are no contractual obligations on the part of either party inferred by the availability of the software.
Its simply a question of getting what you have paid for.
Before you think M$ has to turn over their code... They don't sell their software. And its not written to specification to actually control specific devices in the hands of the government.
New York City and the MTA are screwed because they're morons who got into a shady deal (there's probably some corruption story in there somewhere,) and walked in a field full of bear traps with eyes wide shut. And the strap-hangers are bent over and greased up. [I'm one of them!]
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
>The Village Voice is running a story about New York City's troubles with the vendor who built the subway Metrocard system - magnetic swipe cards that work so well, I almost got arrested once because the system was... deficient.
I looked at this story, read the above, and then realised how silly it was.
Think about it....
If you open-sourced something like the that then you'd open up the market for black-market/fake Metrocards, which would cost the Govenrment lots more in the long term than a few glitches with a contractor.
Face facts. Open Source isn't the be all and end all, and isn't ideal in the Real World.
On a side note, isn't Mandrake funded by Macmillan? They are in the perfect position to sell software to schools. With the recent release of OpenOffice they could profide a complete office automation system for schools. This would also be a good way for students to learn some real computer skills instead of how bad systems behave irrationally and the magic three fingered salute.
-- Remember: Wherever you go, there you are!
http://www.openclassroom.org
...
The site will look sorta familiar
.
Given enough aircraft, all runways are shallow.
But the idea that you can never sue someone for making open-source software is nonsense. The ability to sue comes not from the closed source, but from the contract one has with the authors. In the case of an open source ATC system, one would hire a company to write it and maintain it with appropriate penalty clauses in the contract just as you would with closed source -- the difference would be that if, ten years later, the company that was maintaining it turns incompetent or goes bankrupt, you can take the code to someone else and contract *them* to maintain it; and, of course, like all open-source software, you get the benefit of peer review from all over. Just make the company with the maintenance contract responsible for reviewing the patches submitted before accepting them, and there you go.
;)
Imagine the benefit you would gain from a few dozen air traffic controllers hacking on the code they depend on every day in their spare time, with professional programmers working under bloodthirsty contract standing by to filter out their mistakes. It's a veritable Open Source Utopia.
I'm a consultant working for a company which, for various reasons, always supplies source code to the client. It is the nature of the product we use, but it isn't considered an issue at all.
On the other hand, if it was closed source, well that wouldn't be a significant issue either. You get what you pay for, closed binary, or sourcecode. It's called business. *I* don't like to see a company in the positition where they are stuck having to call me in. I'd much rather they call me in because they like my work, and the company I work for, rather than because they feel like they are stuck between a rock and a hardplace.
But this article isn't about that.
This article was about how the city bought 1500+ machines, and only used 1000. Uhm. Guys, Thats NOT that unusual. No mention is made of how many of the remaining machines are scheduled to be installed. Nor how many will be kept as spares. (Lets see, install 250 over the next 2 months, leave 250 as quick replacements for defective/vandalized machines...)
Doesn't really seem so unreasonable.
The article discusses some of the issues with code changes, updates.
These are NOT to be taken lightly. Its a complex system, it should have some form of control process in place and takes time. Some changes could take 2 minutes if you had access to the source code... it could also take 12 months to clean up the mess if you fuck up. It happens.
I'm not saying this system wasn't more expensive that it should have been. I'm not saying the company was as responsive as they should be. I am saying that there is nothing special about the company, or the situation. (note: many companies will not honour waranties if you modify their code... why? because you may have screwed it up. Not their problem. Its yours.).
(Screw with my code and I might get upset too, especialy if it is an ongoing project.).
How much of a system like this should be open source? Well, how many such systems exist? Almost NONE. the Development of such a system is expensive. Even if the city decided to do it themselves it would cost them a lot of money to develop the software, and the hardware. Someone has to build it, and, most of it is one-of. 1500 machines? For that you don't get great deals on parts. 10,000 card readers and you might start getting deals...
The entire attitude in the article is offensive. (speak-and-spell interface). What would you prefer, mouse driven??
I haven't been to New York, so I haven't used their metrocard system. On the other hand Vancouver B.C. has ticket dispensers for their ALRT system. An idiot could use them... oh wait... thats the idea. Anybody can use them.
This article expresess multiple, conflicting, political opinions. A cheap shot at replacing people at ticket counters with machines, and a push to opensource it. Open source isn't a cure-all.
I remember one cow that would disagree. I believe he was in the Restaurant at the End of the Universe
Cheers,
Steven Rostedt
Steven Rostedt
-- Nevermind
What kind of crieterea is that for mission critical national security software? Lets take a poll and see how many people can use what!. What kind of collosal idiots are making decisions based on a popularity contest. Why not shoose the software that's going to work best, work fastest for the least amount of money and then train your people. Most government labs have insane turnover anyway for chrissake. More tax money spent down the drain.
A Dick and a Bush .. You know somebody's gonna get screwed.
War is necrophilia.
> Custom written software like that is one instance where you *can* sue the people who wrote it if it fails, and ...
It is very difficult to specify software to the degree of detail where you can actually sue someone. Specs change during development anyway.
Suing someone rarely does any good. The software vendors are experts at this - it is a core competency to negotiate contracts with plenty of holes in them that aren't apparent at the time.
And of course the PHB is breathing down your neck wanting to get started and can't understand why you are being so pedantic about the contract. The sales guy is telling him they are keen to go but 'you man' is holding things up.
The situation is not much better than trying to sue MSFT when powerpoint has a bug.
A good vendor will let you have source code. If not it is a good sign you are about to be done over. At a minimum you should have an escrow agreement so that they place the source at a trusted thrid party so it they go bust or - worse - get taken over by CA - you are not left twisting in the wind.
Finally it is no compensation to sue someone if your plane just crashed. (Airline safety briefing for busy people: If the plane crashes you will die).
Imagine trying to explain how to "hack" Metrocards by scratching out parts of it, using ascii pictures in some l33t 0d4y t3xt f1l3. heh.
What a nightmare.
I like the part about "locking down" part of the code to keep "geeks" from figuring out how to put cash on the card. Will people never learn that security through obscurity is no security at all? Make the mechanisms open and the encryption secure and everyone will be happy (although I suspect retrofitting NYC's Metrocard system to be truly secure would be an enormous undertaking).
It's outrageous that we should be in the thrall of a corrupt company like this. I feel that, in many ways, this sort of thing is a much more powerful argument for open source than any individual company's case history.
Just curious: how did you "nearly get arrested" because of Metrocard's deficiency?
How would open-sourcing software work for ATC? First of all, the majority of the problem is the hardware. Open-sourcing the software would not help that one bit! The US by far has the best ATC setup in the world, even with all of it's deficiencies. The FAA has long had plans to upgrade the systems, but those have been fought with budget overruns and slipping deadlines. In short, nothing's going to happen until the current system just stops working one day.
Also, how would your average OSS programmer contribute? In short, they wouldn't. I am a pilot, and I am no expert in ATC procedures and regulations even though I fly "in the system" several times per week. I also know many pilots who don't know a lick about how the internals of ATC work either. Don't get me wrong, we know how they work for us, but not all the behind-the-scenes stuff. Your average Joe programmer is going to know less and I just don't see how they can contribute greatly to improving a system that is broken is more ways than poor software.
I mean, really. Asking Slashdot whether it thinks governments should use open source code? Why don't we ask Microsoft employees if they'd like the DOJ case to be dropped? Or perhaps we should ask the Pope if he's a Catholic?
I worked for the Dept of Veterans Affairs about 5 years ago working on their FOIA (Freedom Of Information Act) CD-ROMs. The VA released all the source code, except for the encryption routines, and would give it to anyone who paid the $5 fee for development and production of the CD (remember this was 1995). Presumably, other agencies that write code also have to release it through FOIA.
-- Ever notice that fast-burning fuse looks exactly the same as slow-burning fuse? I didn't... (Edgar Montrose)
Here's a plausable scenario: 1. Government adopts Free systems (they already use BSD on the most secure networks). 2. Government modifys and improves the implementations for their own use. 3. Government refuses to release new sources due to "natonal security". 4. Millions of users break BSD/GNU law. 5. No one who cares can do anything about it. As long as the Prez doesn't use SlackWare(...)
-----------------------------------------
Perversely greped and groped by PowerPenguin
how does this metrocard thing work again? I'm trying to swipe the card in my butt but it keeps getting stuck. Oh well time to go find my "open source" monkey. I can wipe my ass by sitting on it. Maybe then this metrocard thing will work
Take this personaility test.
Gov't really does it to itself. They can't do the projects internally because they can't justify the head count. These big projects are rare. And let's face it, cities aren't in the habit of laying off people.
So they sub-contract. Generally, they have to take the lowest bid. No one wants to end up on Dateline saying they spent an extra 11 million. This means we don't always get the best solution. Man, and if they try to import in a tested European or Asian system. The shit hit the fan, cause it's gott'a be made in the f*cking USA. Dispite the fact that most of the US's mass transit system is shit. (But I digress)
It's not uncommon to read about a city scraping an entire project and starting over again. We saw a lot of this in the late 80's when cities were big on hooking modems onto water meters, etc, etc.
This is not to say big business doesn't do this as well. I've seen plenty of business men take the aditude that they don't want to be a development house. We'll buy something off the shelf. If it doesn't work, we'll buy something else.
Government procedings occur, for the most part, under full public scrutiny. Why isn't it natural that the software should be as so, also?
Pax Digitalia
If Microslop spent as much money making a reliable, secure product as they spend on marketing, they might actually be able to produce software that DOSEN'T suck shit.
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
As a frequent user of the metrocard system, I've got to give them some credit; it works pretty much flawlessly. Yes, sometimes cards don't swipe, but it usually works fine. And the vending machines are a dream come true; imagine one that accepts every dollar, no matter how crumpled.
As a side note, I'm curious as to what the vending machines are running. The PATH subway system, which connects NYC and NJ has similiar boxes which seem to run SCO (I saw one that had run into a problem at boot time; was kind of funny to see the control-D message in a subway station).
--
The fact is, I've never bought into the two Big Ideas of the OS movement. I simply don't agree with the Stallman argument that software "ownership" is an Extremely Evil concept. (No, I don't want to get into that argument right now. That would be very Off-Topic.) Nor do I agree that the Bazaar is always going to produce better software than the Cathedral.
That being said, it's pretty clear to me that some projects absolutely must be open-source. For one thing, OS software methods do work better some of the time. The Cathedral has dicked around with GUIs for decades and given us the Xerox Star, Microsoft Windows, OpenWindows, and the ultimate in unprogrammable bloatware, CDE. The Open-Source community has been around for a few years, and has almost absent-mindedly given us KDE, GNOME, Englightenment (which I personally find esthetically appealing even though my brain isn't wired to use it), etc.
Even the closed-source Mac is an example of this. Even if you accept all the fancy usability design principles the Mac is based on (and I personally feel that the Mac is overrated in this respect -- benefiting from the absence of competing usability design principles) you have to admit that Apple is doing a lousy job of maintaining them.
A more important consideration is simple security. Bruce Schneier has convinced me that the only way to secure system software is to expose the source code. That enables the user community to verify security claims. The alternative is to rely on the untestable assertions of closed-source developers.
__________
The programs could still be audited and controlled as they are now but patches could be submitted after the initial release, increasing the speed at which bugs are fixed.
Working this way, air traffic control is a place where open source can work, and IMHO work well.
People may not be happy with using ATC which has been built by the open source community (esp if they've used Mozilla) but who could object to an extra few thousand people checking/fixing the code after release?
TWW
"Encyclopedia" is to "Wikipedia" what "Library" is to "Some people at a bus stop"
The FAA doesn't want to change the method by which they do the Air Traffic Control, to open source, or otherwise.
IBM had spent years developing and testing a system of computers aiding Air Traffic Controllers, in place of the handwritten system they use... The FAA put the kibosh on the project, sticking with paper and pencil.
I highly doubt that we'll see open source powering ATC systems in any near future. Open Source has its place, and could find a home in many Government run systems, but due to stick in the mud Administrations, it won't stand a chance in some quarters of the government.
Besides, think about it--- software for flight control must be as reliable as software for NASA-- no bugs. not one. not ever.
This is a wee bit different than the 'release early and often' philosophy.
A host is a host from coast to coast, but no one uses a host that's close
This is exactly the place for some overarching rules in favor of open source. Change is always expensive in the short run. And it is understandable if businesses chose to stick with NT for these reasons. But the whole point of having a government is to be able to include in the decision making processes inputs that markets tend to neglect.
The government is a large enough contractor to actually set standards of interoperability. And had it chosen to base these on open source, it would have benefited immensly the IT infrastructure of thu US and eventually brought productivity gains to the whole economy. Yes, there would have been an initial price to pay-- hiring a new unix sysadmin, for example, in your case. But frankly I cannot see a lot of better uses of my tax dollars than improving the sagging IT infrastacture.
That said, I don't agree that all or even most government software projects should be open sourced. This is a ideological rallying cry with little merit. But government units should not be able to make purchasing decisions in favor of "cheap" products that do not follow a much more stringent set of requirements, including:
use of open source infrastrucure if one exists.
( os, database engine, servers, etc.)
-- look, cheese ahoy!
I think the author of the parent has a point. There is a lack of responsibility with open source, the way we do it with GPL and such. I think there are possibilities for some open source solution, but I doubt the way "we" wrote Linux and such is a good model for mission critical software.
-
I've had enough abrasive sigs. Kittens are cute and fuzzy.
I think this is idea is interesting and would certainly be beneficial. I don't know how the public school system is charged for the software you mentioned (e.g., whether they pay a licensing fee by year or one site license for an unlimited time), so I cannot say whether or not this will actually save any money for existing schools.
However, there would be other big advantages to an open source records system for schools. The most obvious to me is that such software could be made to work on any number of operating systems and hardware models, so that the school need not be bound to a specific platform or company. Additionally, it would be possible to implement other features besides record keeping. (I think it would be beneficial if students could log in over the Web and check their current grades, so that grades on their report cards don't come as a surprise.)
In terms of open sourcing all government-used software, there are many other advantages to open sourcing selected applications, but everything should be done with moderation: while public school software is a great candidate for open sourcing, weapons control systems, as an off-the-top-of-my-head example, are not.
Just my two cents worth...
Nathaniel G H
This reminds me of a survey that I once took regarding household appliances (supposedly 3rd party research): would you be willing to buy product-X which improves your life in so many ways and only costs $19.95 or are you so short-sighted and cheap that you won't even spend 2 hours wages on something this great? I mean, really, was that a question when he (Michael) answered it himself? Most of the time the /. editors try to be at least a *little* fair wrt closed-source software (note the mainly positive stuff said about BeOS). But this is over the top and arrogant. I for one sure as *hell* don't want open-source air traffic control software. How are you going to test it? Not at *my* airport!!! Custom written software like that is one instance where you *can* sue the people who wrote it if it fails, and you can be daggon sure that the people who do the grunt work are quite aware of that, and do a better job than most of the commercial crap that's out there. gah, next thing you know I'll be joining the ranks of people who claim /. is going downhill. I will admit that I remember with fondness the old Multia that used to run everything. That sucker got hot!! Well, I'm off to take my medicine.
proudly ignoring the preview button,
So Germany and France are both very close to either strongly preferring or requiring open source software for certain kinds of government implementations (or so we've heard). OTOH, i've read that most US federal agencies are strongly perferring COTS (Commercial Off The Shelf) software solutions with a minimum amount of integration and custom code written.
This is the challenge that open source actually solves fairly well and fairly directly: COTS products are preferred because government agencies (and most private organizations as well) have a proven inability to develop software (they just write crappy code, manage their projects poorly and usually never finish). Given this environment, they'd like to 'just buy' everything (My slogan is that although you can do almost anything, you can't "just" do anything--integration is tough and no amount of management ignoring it will change that).
On the other hand, we read stories like this (and this one if funny, but hardly unique) about governments getting srewed by the commercial software vendors they use.
Open source splits this right down the middle--you get competent people to develop your products but you get access to the source to make changes if you want. cool, eh? government agencies who are nervous about this kind of thing can take a middle ground of establishing reliance on open protocols and requiring commercial vendors to support them.
Should the government go open-source? In terms of software, yes. Maintenance costs are cheaper when you go in-house (most of the time) and these sorts of positions are somewhat immunized against graft. Hardware is another story: They should use straight standards, but this IS government we're talking about.
In reality, it won't happen. Closed software yields great fiscal rewards for corporate America. They can survive hell freezing over as long as they have that government contract.
I support the Gallic and Deutch moves towards an open source friendly government.
I used to be someone else. Now I'm someone better.
Real life is underrated.
Working as I do at Teledyne Brown Engineering, I see some of the NASA software-development process second-hand. I know they outsource a lot of it--such as the Payload Data Library, which TBE runs for Marshall Space Flight Center. [I wonder if that link is visible by the bulk of /.ers...] NASA is progressing more to industry standards for things--maybe if open-source becomes industry standard, the Feds might follow suit.
There are, believe it or not, some things that won't benefit from being open source. Besides, the Feds will probably decide that some things will create more trouble than they're worth...
--
-- Geof F. Morris
Now, that being said, wherever possible we do get involved in Open Source computing whenever its realistic to do so from an overall perspective. Bruce Perens even came to our site to lecture on the topic (and it was great! Thanks Bruce!!) We recently migrated an important database application from HP/UX to NT, and we seriously considered hosting it on linux. In fact, our tests showed that the db (Sybase) was clearly faster on linux than NT, yet we chose NT. Why? Simple...there are about 25 people in our group capable of learning how to administer an NT box in a reasonable amount of time, there are about 4 or 5 available who have a reasonable background in *nix. It was purely a business tradeoff. As another example, we have many people looking into Beowulf clusters running linux, because that is the appropriate tool for the job.
SuperID
Hardly the oracle to discern the truth. They don't have a NYC bias or anything, ya think???
I bet Cubic has a few interesting tales about the folks in New York, as well.
I liked the part about NYC wanting "one modification" after the project has been planned -- and NYC is p-oed that Cubic has to cut shipment by 400 machines to stay under budget.
Modifications to fixed price contracts have to be paid for somehow. Should the company pay? Why?
Most contracts are written such that the buyer gets all specs, software, spares, etc. when the contract terminates.
So terminate it, NYC, get yourself another whipping boy. Force them to lowball you to get the contract, then ask for mods when they're tooling up for production.
Sheesh. This article is just typical NYC whining about their own frickin' mistakes. It would be comical if it wasn't so sad.
Treatment, not tyranny. End the drug war and free our American POWs.
See my user info for links.
Open source does not mean that any code submitted gets added. I, for one, would be pleased to know that thousands of talented programmers around the world could review code responsible for the safety of so many travelers. Isn't this exactly what we're all demanding for Carnivore?
Develop a standard school administration file format (XML...) and some free applications that can run on top of it, for teacher's gradebooks, district records, management, & reporting, etc. Make it open source... heck, get [insert government entity] to pay for it's development.
Think of how much the school systems could save, if each district didn't have to pay millions for crappy on-off, outdated applications from houghton-mifflin or macmillan or whatever...
My wife used to teach highschool here in Texas. The systems they had for teacher's gradbooks (required, mind you) were horrendous. DOS-based, for starters, always breaking down ("submit 2 floppies, and 3 printouts, just in case"). A huge waste of time and money...
I'd love to see open source make headway in this area. I'd love to see some gov't money going to fund it...
---
... is to do what the DoD does on most of it's coding contracts: require that the contractor make its source code available to the DoD. That is, the contract requires that the DoD has clear rights to use the code and have others work with it, but the contractor retains copyright control.
Honestly, for political as well as institutional reasons, I think this is the best way to go for quite awhile. All coding work (and all systems that have any sort of code, from EPROMs to Java) should have the stipulation that the gov't has a clear right to use and modify the source code with no additional compensation. Now, that doesn't mean that the gov't can sell the code or give it away under a OpenSource-style license, or even that they can reuse the code for another project (that should be negotiated in the contract, with appropriate compensation for the contractor). It simply means that the gov't can get others to fix problems with the code if need arises (big holes, the contractor refuses to finish the job, the contractor won't maintain the code, contractor goes out of business, etc....)
This strategy is kind of the minimum resistance path - it still protects closed-source vendors from having their code released to the public (and let's face it, you wouldn't ever see mass-market software (and its low price) make it into gov't use if the companies were required to give away the code to anyone). And it saves the security debate for another time (that is, the debate over Open vs. Closed software won't impact the decision to require source).
Personally, I'd like to see the government start requiring perpetual license for all code to be used internally in any way. They'll be a lot of resistance for this, but I think there's a much better chance of this happening than if the gov't tries to require a GPL/BSD/whatever code license for all work. If you do that, then the gov't loses all access to mass-market software, and it becomes a completely custom job.
On second thought, if it all requires custom programming, well, hey, Welfare For Programmers! <grin>
-Erik
There are always four sides to every story: your side, their side, the truth, and what really happened.
Asking Slashdot if the government or anything else should be open source is a lot like asking a cow if we should be vegetarians.
Thats right. In cubic's infinite wisdom, they made metrocards as unforgable as possible by making the card themseves analog. This turns out to be really stupid, because the machines then have to expect slightly mangled data (after all, the cards stay in my wallet next to all my other credit cards.) So, if you scratch the right portion of the card, than you can stay under the machine's error threshhold while destroying the part of the card that holds the current price (Apperantly the original amount and the current amount are stored in different locations on the card).
Stupid MTA.
Well, that will show them for putting cops in every station to arrest me for truency after 3:00 and during the summer.
--Alex the very bitter Fishman