Open Code in Public Procurement
mpawlo writes: "I wrote something on public procurement and open code that you might want to share with your readers. In my opinion, it is time that public bodies and governments look over their public procurement policies to warrant competition. I don't think free software or open source should be the only choice when it comes to public computer programs, but as of today, public bodies all over the world designs their requirements in a way that rules out all Free Software and Open Source alternatives already at the drawing table. May the best computer program and license win! That's the only way to get an effective allocation of public money when it comes to public computer programs. Maybe a good topic for discussion among Slashdotters?"
Closed source programs like Office are the standard in government (just like in the private sector).
:: I'm The Man Now, Dawg!
When you have committee after committee that needs to share data, it's beneficial to use one standard so that you spend less time haggling with technological problems and more time on the real work -- governmental issues.
As nice as it would be to see all governmental organizations, agencies, committees, etc. using open source or free software to get their work done, it's not something that will happen overnight (or even this year).
But if the coders keep coding and the zealots keep shouting, they'll hear the voice of reason. Perhaps even just the monetary issue is enough to get them all to switch over.
EricKrout.com
If you celebrate Xmas, befriend me (538
Could have something to do with the fact that vendors often supply "helpful" templates to procurement officers that are often so tightly written that only one vendor's product will meet the requirements, much less that any open source product would.
One CPU cycle wasted on digital restrictions management is ONE TOO MANY.
I'd like to see some discussion on open-source models that give the user freedom to choose their upgrade (libre) and gives programmers the right to fork, all while not requiring software to be free of charge (gratis). For more thoughts, see Distributed Copyright.
At my university (Nebraska-Lincoln) we are currently facing budget cuts to the tune of something like 8.3 million dollars. Now, the university has a contract with Microsoft to for about a few million a year to supply all computers on campus with windows/office.
When somebody suggested not renewing the contract (Thereby saving a few mil) and instead switching over as much of campus as possible to Linux they where laughed out the door by the ITS people. They said, among other things:
1.)Cost too much to implement (retraining users, etc)
2.)Would be too hard to support
3.)Wouldn't provide students with the knowledge of computers to succeed in the real world I.E. Microsoft software is used by 99% of the business world and having everything run linux would simply not be effective in teaching students how to use 'real world' applications.
Where they right? I don't think so. But instead of cancelling the contract they are now cutting faculty raises, a number of teaching centers, and some extra programs.
Before we go and change how gov'ts contract software we must realize just how damm impossible it is to get them to get past microsoft's FUD.
Government doesn't have a reason to change, therefor, they won't. One of our biggest clients is a government body. They've been ingrained for a long time with Microsoft and just aren't going to switch from MS Office to say, Staroffice. Reasons for this are actually valid.
1) They are comfrotable and familiar with Word and Excell.
2) Some of their Access documents would be hard to render properly in Staroffice.
3) They've invested a ton of money into several Visual Basic programs that use Access as a back end.
As long as government agencies, departments, etc need things like that (which they've spent alot of money on to impliment), they are loath to switch.
Slackware forever. Honestly, what else would you trust when it absolutely positively has to be stable, secure, and easy
You haven't quite understood the open-source and free-software business proposition, have you ? What's more, in my views, when the government buys Microsoft software, it makes them a little richer and that threatens my job in a small non-Microsoft company that much more. How about a little of that ?
"The government should always choose the best computer program and IT solution at any given period of time."
You forget half of the equation : a government is more than a company, and they have to take national interests into account, which is usually more important than the technical solution. For non-US governments, that often means one of the most important requirements is to not run closed-source software from a US monopoly.
One reason I can see is public confidence. The public probably thinks that open source means hackers , knowing how the system works, could easily break into the system and steal whatever they want. Open source simply does not have the reputation of reliability that most proprietary systems do simply to the fact that most people automatically assume that anything "professionally" designed by a company for the purpose of making money will by automatically better than something made by people in their spare time.
I stole this Sig
Well, "open code" could mean several things. If it means free software, or Free Software, there is no procurement at all: people inside the government just use it; there is no procurement. By the time procurement happens, most likely, the government IT people have already ruled out free software. Procurement then involves two sides: the government and vendors. The vendors could, of course, involve companies like RedHat.
You also have to weight the costs/benefits of open-source, and what kind of support contract(s) you can get for such systems. A system can be completely open-source and proprietary at the same time, making the learning curve for any potential support personnel unnecessarily steep. Though Apache/BSD shoud be pretty much standard everywhere, as I can't see a good reason to use a Windoze webserver in a minimal cost environment, such as government. Then again, the government never seems to have any incentive to make fiscally-sound choices, as what profit motive do they have?
The biggest trick the devil pulled was letting lawyers become politicians so they can write the laws.
IMHO, this idea is both reasonable and constructive. It's certainly not the radical pseudo-communism the purveyors of FUD will inevitably make it out to be.
... risks none of the dangers of strategic behavior that closed code, or controlled networks, do. If open code is used strategically, then the resource to counter that strategic action are always available. Innovators can rely upon the promise of open code in their innovations. They need not worry that what they develop will be swallowed by the platform they develop for.
Personally, I like the way Lessig put it in _The Future of Ideas_ when he argued that the government should encourage the development of open code.
"Open code
This encouragement should not be coercive. There's no reason to ban or punish proprietary providers. People should be free to develop code however they wish.
But a government has its own interests, and closing its resources to others is not one of them. If the federal government develops a system to handle welfare claims, what reason does it have for hiding the code for that system from the states? Why not let the states take that code and build upon it? And if the states, then so, too, with the universities. In each case, the aim should be to expand the reach of these powerful and valuable resources, not to contract and hoard them when to value to the hoarding exists." The Future of Ideas, p.249
Now, the more interesting provisions in government purchasing usually read like this (1) you have to certify that you never have and never will sell it to anyone else for less than you charge the government, (2) it has to be accessible to users with various handicaps (eg visually impaired, carpal tunnel victims, etc).
No-one would specify an OS / Platform or any other detail that is not necessary. A good specification in an RFQ ( Request for Quote) or an RFT (tender) outlines essential & desirable criteria. By specifying outcomes rather than details the market is opened up and the benefits of an open proceurement policy are seen - including $$$ savings , quality and probity.
The problems faced when writing specs are legacy systems & applications - staff knowledge & training included. It may be necessary to specify exact hardware / software ; you may be able to specify "compatibility" requirements rather than exact products ; or in the best case you can write an outcomes only based spec.
It's horses for courses though. The more open the better generally - if you start closing your specs you start removing some of the benefits - $$$$ , probity for example. But , if you have to run the app on an NT4 box with IIS then spec it that way.
If you need compatability with Office apps then spec it that way.
If you need Office then just order a copy. Once you water down your spec by being too tight you may as well just buy the product you want. There is no point issuing an RFQ if there is no market to test. And sometimes this is the best option to take.
MS should open source 3.1, that would be pretty cool just for the sake of it
Whilst I think that forcing MS to open source win 3.1 would not be good thing, I beleive that at the point that a company is not willing to sell/support old software, the said software should loose it's copyright status.
Just like trademark law, if you don't fight to maintain your trademark, you loose it.
AFIK, not so. Having worked for in the public sector a number of years on pretty-much visible (to the public) software, what kept open-source far away was:
1. Middle-level managers aren't sensibilized to try to reduce cost using Free (beer) software; and
2. As soon as the subject arise, higher-level management are afraid to have only one person on the planet that would be able to support that infrastructure. They see MSCE or Novell-certified technicians rain everywhere, but most often only inexperienced people brag about their Unix skills. And if they do, people associate that skill with data centers, not with acting as sysadmin for a small office.
For what I saw, management doesn't care about anything else. For them, if the software doesn't cost anything to buy, then it must cost a fortune to use.
C.
C.
Not me. Open standards are what is needed. That way anyone can write software that interacts with other peoples protocols or file formats.
With open standards, governments (and individuals) are truely free to do whatever they like. They can throw their (our?) money away overseas to multi-nationals, purchase from local closed source developers, use free software or develop in house.
Personally I find the fact that democratic governments are letting private enterprise (rather then their electorates) dictate policy to them, well, ... typical, but disappointing
Honestly, I like the KDE desktop better than Windows. And it gives me a good feeling deep inside to run software that is freely given away, rather than bought. And I do have big misgivings about supporting a company financially that has abused monopoly power to put people out of work and ruin careers.
But the problem is that installing free software is a royal PITA. Currently, I'm on my third install of Linux on my laptop. Now keep in mind, I've been running Linux on my servers for several years now, so I know quite a bit about the system. It turns out that RedHat doesn't support my LCD very well. In the past, I could get around this by altering the XF86Config file, but with this last install, I have painfully become aware that this hack no longer works. I went so far as to remove XF86Config from the system entirely, and X still comes up. So now, I will spend several more days getting this system up and running under Linux.
But guess what? This isn't even an issue with Windows! I don't have to know what hardware I'm running (and don't give me any flames about how a user should know his hardware - it's not like I can bust open my laptop and read the numbers on the video adapter chips - nor should I have to, for that matter). A Windows install will take about an hour, and it works on almost every machine. In a few hours, I've got something that's usable (albeit buggy and insecure), and compatible with the rest of the world.
There is a serious gap in understanding between Windows coders and Linux coders. The Windows folks write software to be used, whereas the Linux folks write software to be configured, as if the joy of configuring software was an added feature. In Linux, I have to know the most intimate details of my machine; not only do I have to know what I want to do, I must know how to do it. Windows makes it simple - I just select what I want it to do, and it figures out the details.
What the Linux zealots fail to recognize is that Linux is not even useable without a great deal of technical knowledge. The reason why many companies have been slow to adopt Linux is because these companies aren't in the business of installing operating systems. They exist to get work done with their machines, and from their perspective, Windows works - it requires little technical knowledge (low training costs), and is compatible with the rest of the business world.
Imagine the following scenario: You are stranded on a desert island, and discover a cellphone, modem, and boxed laptop with two CD's - one for Linux, one for Windows. You open up the laptop to discover that the battery is charged, but alas, no operating system is installed. What do you do? Install Windows, and send out an email for help? Or do you try Linux, and risk finding out a few hours later as the battery dies that it doesn't support your modem?
I like Linux better than Windows. But until the free software community does something about the abysmal installation and configuration process, Linux will only be used by geeks. An operating system shouldn't require that a person learn the equivalent of a bachelor's degree in computer science in order to use it. What I think is saddest about the free software movement is that they have engineered their software so that the common user cannot use it - an intellectual elitism, if you will. If we in this movement were really altruistic, we'd write software that the average user could install and use. After all, why shouldn't everyone use computers?
Correct me if I'm wrong, but wasn't it Linus that said that if MSFT started creating OSs with proper standards in security, stability, etc, it means that we have succeeded?
When the faculty are empowered - and I'm refering to the empowerment of cold, green, cash (for swimming in) - they can do as they please; which, when it comes to computers, is generally whatever their post docs, grad students and senior techs want (I'm a biologist.)
Often, at least here at Columbia, that means individual labs will just go out and buy a bunch of Intel machines on their own initiative, and put Linux on them. This has been going on for the past year or more, and Linux is starting to gain credence with the administration.
At a school where people don't have that kind of funding, individual groups don't have the resources to investigate, not just Linux but new avenues of procurement generally. As is too often the case, if you don't have the sugary wampum to evaluate the different vendors/solutions, you end up stuck with a bad deal.
Anyway, this is a problem that Linux people, especially those at academic instistutions, ought to be pursuing - I say this without bothering to look and see who's pursuing it and how, hoping that someone already familiar with the situation will respond by posting details.
Also, I have a sneaking suspicion that ill-will on the parts of some CS faculty towards some other CS faculty may be hampering the adoption of Linux by certain institutions.
The good and new comes from no quarter where it is looked for, and is always something different from what is expected.
The costs of retraining, cutting over, and so forth for even a 100 person organization would likely come to more than $40K
That comes to $400 per person. Thus, if switching to GNU/Linux would save at least $400 per seat in royalties, GNU/Linux would have a lower total cost of ownership.
Will I retire or break 10K?
I can't think of one secretary that knows what a command line is,
But it's fine for Windows to puke up CPU registers and stack frames when programs crash at them?
Or they should have to know about such things as defragmenting disks
There are plenty of ways in which Windows really is "too technical" for the average user.
In a government environment, how many secretaries do you know that play around with the settings in Windows?? How many know how to change screen resolutions? Most of these functions are locked out anyway!
Indeed there is a whole set of third party addons specifically to deal with the "enduser is the admin" misfeature of Windows.
In a corporate environment, things are controlled and setup via the sys admins. Setting up a user with KDE (or whatever), office package, and they should have little trouble using the system.
One big problem with a lot of more recent GUI apps, especially ports from Windows or attempts to clone Windows apps is that centralised configuration is ignored.
Traditional unix apps would have a central config file and per user configs, some of the more recent stuff only has per user. Even worst expecting end users to set things up, everything from web proxies to huge forms which Star Office comes up with...
But guess what? This isn't even an issue with Windows! I don't have to know what hardware I'm running (and don't give me any flames about how a user should know his hardware - it's not like I can bust open my laptop and read the numbers on the video adapter chips - nor should I have to, for that matter).
Guess what there are plenty of anecdotes about Linux just working, whilst Windows required all sorts of fun. Starting with the old favourate of trying to install drivers for a CDROM drive from a CDROM...
Knowing what the hardware is is a matter for the installer/maintainer. Which isn any corporate/government type setting is not the user.
the UK's commercial wing of DERA (Defense Evaluation and Research Agency) produced this report: QinetiQ_OSS_rep.pdf [govtalk.gov.uk]. Which is the most pro-OSS report I've read.
Considering what these people do "Blue Screen of Death" takes on an all together more serious meaning...
IIRC DERA did evaluate some Windows based battlefield systems, not sure how they actually performed.
Companies pay a good portion of the taxes that pay for publicly-funded software. Governments should be encouraged to fund/use/get software that doesn't exclude use in closed-source software.
On this basis they should probably automatically exclude Microsoft, since they are expert at tax avoidance...
companies aren't in the business of installing operating systems. They exist to get work done with their machines,
We used to have one "help desk" man that spent more than 20 hours/week just re-installing Windows, Office, etc., on user's desktops as the 100 users here kept developing problems for which MS tech support's recommendation was a disk wipe and re-install. I'm not talking about upgrades here, but re-installs. Win 98SE is rather more stable, 2000 is said to be better, but still, Windows is not a great solution for those that don't want to be in the OS business either.
Granted, Windows gives a complete ignoramus about a 90% chance of getting through an install on recent hardware without ever reading the instructions, while with Linux you've got to know what the hell you are doing for the initial install. But AFAIK once Linux is running right on a desktop, the only reason to ever re-install is to upgrade with new features. If you've got a few hundred desktops, maybe you could replace several guys who don't know much beyond "stick in the CD and run d:setup.exe", with one guy that actually knows how to solve the problems with Linux setup. (Servers also need constant security upgrades, just like Windows -- but does Windows still have to be re-booted for every patch? And you'd damn well better have someone knowledgeable maintain your servers, no matter what OS is used.)
I do know the Navy has put Windows into some ships. And they had the main propulsion BSOD'd at least once...
This has been driven by the EU recommendation to consider open source mentioned in the past on Slashdot.
Main body are these recommendations:
BTW: PLEASE DON'T SEND COMMENTS TO GOVTALK if you are just going to say 'me too'. The 7 comments that are there are all pro-open source and we dont want to sound like fanatics, do we? Remember how a lot of the comments to the EU on patent law were essentially ignored for this reason - don't duplicate the arguments of others.
Disclaimer: I don't work for the UK government, but I write software that gets sold to them. Which includes a lot of open source stuff. I just happened to be reading that policy today before I read /. ...
Cheers, Baz
It's the capitalist troll again. Is this the same guy, or is there a group of Randites out there who do this?
Big hint: The US isn't a capitalismm.
As soon as the government starts passing laws regulating business you don't have a free unencumbered capitalism, you have a socialism.
The government already passes laws which benefit Microsoft (that their stupid restrictions in the EULA are binding in some states) so why shouldn't they pass laws which help other companies too?
As soon as the government starts passing laws regulating business you don't have a free unencumbered capitalism, you have a socialism.
Actually they are not talking about laws regulating what software other people can use, but regulating what government departments can use. Whole different kettle of fish. The problem is that the Government is such a big organisation that standardisation on an app creates huge ripples in the industry (plus any other businesses that want to be compatible with the gov.)
From J.D. Hogg's comment:
You forget half of the equation : a government is more than a company, and they have to take national interests into account, which is usually more important than the technical solution. For non-US governments, that often means one of the most important requirements is to not run closed-source software from a US monopoly.
I would say that a government is not a company in the usual sense of the word at all. Their primary interest is to get their administrative jobs done, and for that, the best technical solution is paramount. [Spurious example] Would you like to hear that your tax refund check is going to be an extra month late because the tax department computer system purchased was inapropriate for their needs because they couldn't find an open source vendor that could deliver the goods dispite closed source companies that could?
I too belive that open source should be using in government, but I don't think it should be jumpped into and mandated across the whole of government when it is inapropriate. Moderation in all things.
For government purchasing of IT solutions, cost effectiveness and price/performance is more important than vendor relationships or IT community placating. The various government departments have their own jobs to do and they should use the best tools at their disposal to do it, regardless of where they come from. It is up to the linux community and the companies that participate/support it to supply the best tools for the government, so that choosing open source will be a quality desicion and not a philisophical/political one.
"I'll take the red pill. No! Blue! AAAaaaahhhhhhhhh"
- Monty Python meets the Matrix