Even so, the guy still walked out from SCO with some intimation of what the code actually was. Assuming it didn't look ridiculously generic and therefore likely to be produced by 2 people working on the same problem at the same time, it would have some defining characteristics that a grep through the source would probably find pretty quickly.
If it's in an old version of the kernel, then it may just take a bit longer to grep for it.
I'm not talking about something as gruesome as a SCO copyright message, but you'd think there would be some aspect of the code that would remain in the guy's mind when he left SCO.
Furthermore, I can't believe the individuals who signed the NDA were that disinterested that they didn't do exactly this. I know *I* would have done so, if only to satisfy my own curiosity.
Assume just one of these people did this, and now has concrete proof (if only in their own mind) as to whether SCO is talking out of their arse or not. Would that person not be inclined to find a way to make this information public, via whatever obfuscation process was required to ensure their identity was untraceable? I mean, these guys were picked because of their kernel and/or coding knowledge; surely they could make a few discreet enquiries about anonymous mail relays and how to use them...
I'm not encouraging these guys to flout SCOs NDA; I'm just surprised that it hasn't occurred through "osmosis" as normally happens with NDAs, and there's not a general feeling among OSS people that the problem has gone away if it actually ever existed.
Several weeks after people were able to view the supposedly offending code (under NDA), there's still no report of what it actually is.
This is in contrast to every known FUD convention, where it's normal practice to sign a NDA, look at something secret, wait a few days then quietly have a word in your buddy's ear and get him to post some still-speculative-but-extremely-specific detail of what it is you looked at.
Why the sudden maintenance of SCO's secrecy, when there's an industry-wide history of violating similar NDAs at the first opportunity? How can we not know even the tiniest specific detail of SCO's case, yet we know e.g. details of every close-kept Apple product release several days before Steve Jobs announces it?
A few weeks back, I honestly expected the following to happen: - a few people sign the NDA and view the code in question - (nothing happens for a few days) - new code gets quietly released for functions A, F, H and Z in the kernel, gets exhaustively tested by several key Linux people and very quickly appears in the next kernel release - confident pronouncements from Linus, RedHat, SuSE etc. that they are absolutely sure the SCO case has no merit, that they believe (but can't confirm) the code in question is "old code no longer in use" and so on
Actually, maybe this happening now and I should keep quiet about it. If so, could someone tell me which step we're up to? I promise not to tell
Under the present IP and patent laws, he's probably right - given MS holds zillions of patents, it would be very surprising if some Linux stuff didn't step on these patents.
However, I don't think he'd be silly enough to try to enforce these patents. Given IBM's love and support for Linux, and given that MS is almost certainly stepping on some of IBM's even larger set of patents, then he'd potentially be setting MS up for a world of hurt.
There's been a gentleman's agreement for years between IBM, MS and many others that sharing their patents via cross-licencing agreements is the only way to advance the industry as a whole. If MS tries to go against this agreement, IBM could give them a world of hurt (and get a lot of pain in return).
Since when did the guys who run jails start using their imaginations? Maybe they need guys like me to start helping them out, coming up with wild and wacky ideas that would get prisoners roaming free again.
I just had a vision of releasing prisoners early, and using this balloon/camera thingy to track their every move. It hovers over them all the time, and feeds pictures back to some central point so their whereabouts can be monitored at all times. They're free to go wherever the rest of us can go, but they have this camera hovering over them all the time until their sentence is up. Think of the savings in jail accomodation!
Then the fatal flaw hit me: the ex-con goes for a job interview, holding a 1m balloon with a camera suspended below it, on a piece of string, in an office trying to describe how he'd be a great employee.
The local bad guys' public bar would look like a fairground, full of shiny balloons. "Mum, can we go play in that new inside park?" would be the cry from the kindergarten set.
Or imagine a typical NBA game. With the number of balloons that would be floating over the players, nobody would be able to watch the game. Hold on - there's no reason the balloons couldn't contain advertising.
Well, actually that's several fatal flaws, but I still think it has "weird and cool" merits that override the "it's a really, really dumb idea" issues. This idea has a really great application somewhere, but I just can't see it at the moment.
Putting together the responses to date, it seems like this is the story:
SVG: lots of potential; the ideal solution if the plugin is installed or easily installable; XML based
Flash: the pragmatic choice; installed nearly everywhere; works reliably on all platforms; can be driven via XML
VML: good fit for IE 5+ browsers, since it's supported in a default install; unsupported in Mozilla; deprecated technology; XML based
Java applet: slow to startup; JFreeChart looks very powerful provided startup times are acceptable and browser support is available
In case you can't tell, I'm facing the same problem as the original poster. I also can't deploy server-side solutions, so PHP, Perl GD, etc are all out.
Based on this, I'm inclined to build a generic XML solution, then rely on plugin detection and XSLT to deliver either SVG or VML as appropriate. That covers me for IE 5.5+ and anything with SVG support. The largest group that I don't have covered is Mac users and people with older browsers, and I could probably come up with a different XSLT to cope with them if the demand made it worth doing.
When you say OSS software isn't quite ready, and from the context I'm assuming you're talking about MS Office replacements and similar end-user-facing stuff, you have to remember that governments aren't full of people creating complex spreadsheets and Word documents.
Many/most government employees are "process workers"; people who use a very small number of programs (e.g. a Web browser) to perform largely repetitive tasks. There's very little knowledge or IT training these people need to do their jobs; what they need to know is (a) how to login, (b) how to start their application, (c) how to navigate through the screens and enter in data. They *don't* need training on KDE, Gnome, Unix file system layout etc.
For these people, I'd say OSS is well and truly ready, and has been for some time.
On that basis, modchips can't be illegal per se in the US - they're only illegal if they're used (as you say) to do things like play pirated games or develop emulation to let non XBox hardware play XBox games.
How, then, did MS manage to get that mod chip vendor (name escapes me...) to close down a few months ago? Sure MS is a big bully, but the vendor was in Asia somewhere so the DMCA wouldn't apply. I suppose MS could say "we'll take you to court till you're broke", but when there's no question of any valid law being broken (either in the vendor's country or the US) surely the case would be dismissed as frivolous, and MS would have to pay costs *and* reparations to the mod chip vendor (WHAT'S THEIR DAMN NAME??).
Am I missing something? If I was the chip vendor in that situation, I can't see why I wouldn't take MS on in court; I'd put up with a lot of legal pain and the likely loss of my business if I thought there was a good chance I'd be able to make a large chunk of loot out of reparations at the end of it all. Furthermore, finding a legal firm to bankroll me on a "no win, no fee" basis might be pretty easy if I could offer them e.g. 50% of the final payout.
> In any case there is a specific clause in the DMCA > that allows for reverse engineering for the > purpose of interoperability.
Warning: non-US citizen asking...
Based on this, could you then buy and mod an XBox, connecting it to your sound system as a media player system? If you have all your e.g. OGG files stored on a regular PC that runs as a file server (as I do), and play them from the modded XBox attached to your sound system, then you would have used the modded XBox solely for interoperability purposes.
I'm on the point of building exactly this setup in Australia now; would it be legal in the US on the basis of this interoperability clause?
They'll have lots of legacy Windows apps that will need to be progressively ported to Linux, so it makes sense to do this in a planned, application-by-application manner rather than go for a 'big bang' approach.
Furthermore, maybe only 50% of their users will need to run Windows/VMWare at any one time, so they may only need to buy half as many Windows licences as they have machines. Over time, this would decrease, as more and more apps get moved off Windows and onto Linux, and the number of users that need Windows decreases. And before anyone says "this is illegal", remember that we're talking about Germany, and they have much stronger laws restricting licence tying than they do in the US. I'm not an expert in German law, but this is quite likely to be "more legal" (if that term makes sense) in Germany than it would be in the US.
> Yeah, SVG is technically a W3C standard, but it's > hardly a standard in actual web development.
Think about what ubiquitous SVG would do for browser based apps. You could do things like have graphs being updated from real-time data sources, with all sorts of interactivity possibilities. This *alone* is a massive step over what passes today for interactive data presentation and analysis capabilities, and has very big implications for all sorts of scientific and financial applications.
I'm aware you can do some of this using PNG, GIF or JPG pictures, but you don't get anything like the interactive capabilities that SVG makes possible using these formats.
*Finally*, I can start saying SVG is going to be supported natively in a browser, and pushing through projects on that basis.
Until now, I've had to say you can use IE, then get an addon from Adobe. "What? Why doesn't MS support this SVG thing natively? What if Adobe decides to drop support for SVG; then what happens?..."
This is the best news I've read on Slashdot for a while
Does anyone know whether the govt guys that were involved in making the original agreement are still around? If so, are they free from legal action, or can they simply wipe the slate clean and get away with it?
If it's a totally different bunch of guys running the govt now, then they could conceivably justify it as being a dumb decision to enter into the agreement in the first place (using whatever logic they care to present...). If not, then maybe there's some negligence issues that can't be simply signed away with the swipe of a pen.
For those dollars, you aren't in the target market for a.NET solution.
You're not going to have much change out of $20k by the time you buy your server licences and hardware. You've either got to get more loot, or look for another solution (e.g. OSS-based). It's fine if your customer wants to have a.NET solution; you just need to enlighten them that it'll cost more than they've currently given you to build it, so they now have to choose between spending more or having a non-.NET solution.
I frequently find myself in the same situation, and have switched to using Python wherever possible for the Q&D solution.
I've found that, when I write Python code, I generally get the structure of the code pretty right, though I may wind up skimping on documentation, may code stuff inline that should be separated out into functions, and so on. The step of converting Q&D code to production quality using Python is a whole lot smaller than in e.g. C.
Even if the final version has to be in C/C++, then I find implementing a Q&D version in Python first allows me to write better C/C++ code later on. Again, I suspect it's because using Python enables me to get the basic structure of the code 95%+ right on a first attempt, and then it's largely a matter of translating Python to C/C++.
Whatever you do, don't try creating a Q&D version in C/C++ if you can possibly avoid it. C/C++ code tends to be degenerate into unreadable/unmanageable crap much quicker when you're using it for Q&D code; they're great languages for production quality code, but just not that well suited to Q&D code.
I've worked in many government departments, always as an independent contractor. My experience is that government databases, and the apps that talk to these databases, are generally a mess.
I think there's a few issues involved: - a lot of the code is written by people with skill levels that wouldn't be accepted in the outside world. I've got no doubt that there's highly skilful coders working for government, but there's a lot of duds as well and often your code is only as good as the weakest coder in the team. Furthermore, a lot of the weakest coders are the designated "experts" for legacy infrastructure, such as databases, and you find yourself having to rely on their input far more than you'd like - a lot of government stuff has been outsourced, and generally the outsource partners seem to be less diligent for government contracts - governments are forced to send everything out for tender, and may need to give work to the lowest bidder or face internal inquiries. In the private sector, you get to eliminate a lot of tender responses, either because you know the respondent is incompetent from past experience or the price is so low that the respondent couldn't possibly deliver a quality service; in government, you don't have the same luxury of filtering out the crap - in the private sector, a lot of mainframers moved across to Unix, then Windows, in response to changing demand. While these people may not be expert in all 3 areas, they at least know enough to be able to hold a technical discussion with someone else from another area. In government, a lot of mainframers have been doing the same job forever, and you need to tie everything back to their narrow view of the world to get anything out of them whatsoever. You'd better not try to talk "Web Services" with these guys... There's a level of financial support and job security for these types of people in government that allows them to keep doing this, that simply doesn't exist in private enterprise
Now that I've pissed off everyone working in govt IT, I have to reiterate that some of them are extremely good. The issue is that the system doesn't work on a "survival of the fittest" system such as applies in private enterprise.
>...objects written in one language are easily used > in every other language, so you can have 1 > developer using Perl, another using C#. Try that > in Java. Try any cross language development in > Java
You can't compare.NET with Java in this context; you need to compare it with the JRE.
I can and have written Python code that runs in the JRE, using Jython (a 100% Pure Java implementation of Python). It lets me mix and match Python and Java code, accessing objects across languages, and leverage the best of each language. I know you can run a LOT of other languages under the JRE; the number of available languages is at least comparable with those supported under.NET
Disclaimer: ASP.NET is the one piece of.NET I haven't used to any great extent. I have used ASP a fair bit though
Serious question: How does ASP.NET encourage good architectural practices?
As I understand it, ASP.NET still suffers from the fundamental architecture problem of ASP in that it encourages you to mix presentation and business logic. This is a nightmare for design and ongoing maintenance.
Slashdot Mirror
"I'll just confirm that with Chief of Staff BAINWOL"
o rd
"Look out, here comes BAINWOL"
I'll hazard a guess that having all your underlings laugh whenever they hear your name would tend to undermine your authority somewhat...
Is this a real name, or something made up? He's not so much an enigma as an abbreviation:
Bastards
Allied
In
Neutering
W
Of
Law
Even so, the guy still walked out from SCO with some intimation of what the code actually was. Assuming it didn't look ridiculously generic and therefore likely to be produced by 2 people working on the same problem at the same time, it would have some defining characteristics that a grep through the source would probably find pretty quickly.
If it's in an old version of the kernel, then it may just take a bit longer to grep for it.
I'm not talking about something as gruesome as a SCO copyright message, but you'd think there would be some aspect of the code that would remain in the guy's mind when he left SCO.
Furthermore, I can't believe the individuals who signed the NDA were that disinterested that they didn't do exactly this. I know *I* would have done so, if only to satisfy my own curiosity.
Assume just one of these people did this, and now has concrete proof (if only in their own mind) as to whether SCO is talking out of their arse or not. Would that person not be inclined to find a way to make this information public, via whatever obfuscation process was required to ensure their identity was untraceable? I mean, these guys were picked because of their kernel and/or coding knowledge; surely they could make a few discreet enquiries about anonymous mail relays and how to use them...
I'm not encouraging these guys to flout SCOs NDA; I'm just surprised that it hasn't occurred through "osmosis" as normally happens with NDAs, and there's not a general feeling among OSS people that the problem has gone away if it actually ever existed.
Several weeks after people were able to view the supposedly offending code (under NDA), there's still no report of what it actually is.
This is in contrast to every known FUD convention, where it's normal practice to sign a NDA, look at something secret, wait a few days then quietly have a word in your buddy's ear and get him to post some still-speculative-but-extremely-specific detail of what it is you looked at.
Why the sudden maintenance of SCO's secrecy, when there's an industry-wide history of violating similar NDAs at the first opportunity? How can we not know even the tiniest specific detail of SCO's case, yet we know e.g. details of every close-kept Apple product release several days before Steve Jobs announces it?
A few weeks back, I honestly expected the following to happen:
- a few people sign the NDA and view the code in question
- (nothing happens for a few days)
- new code gets quietly released for functions A, F, H and Z in the kernel, gets exhaustively tested by several key Linux people and very quickly appears in the next kernel release
- confident pronouncements from Linus, RedHat, SuSE etc. that they are absolutely sure the SCO case has no merit, that they believe (but can't confirm) the code in question is "old code no longer in use" and so on
Actually, maybe this happening now and I should keep quiet about it. If so, could someone tell me which step we're up to? I promise not to tell
Exactly - IBM will have patents on things that we all take for granted, like compilers, disc drives, RAM and so on.
There's absolutely no way MS can take on IBM in a patent war, and Gates didn't get where he is by knowingly entering battles he can't win.
Under the present IP and patent laws, he's probably right - given MS holds zillions of patents, it would be very surprising if some Linux stuff didn't step on these patents.
However, I don't think he'd be silly enough to try to enforce these patents. Given IBM's love and support for Linux, and given that MS is almost certainly stepping on some of IBM's even larger set of patents, then he'd potentially be setting MS up for a world of hurt.
There's been a gentleman's agreement for years between IBM, MS and many others that sharing their patents via cross-licencing agreements is the only way to advance the industry as a whole. If MS tries to go against this agreement, IBM could give them a world of hurt (and get a lot of pain in return).
All you'd have to do is make the balloons out of something pretty tough. Steel reinforced concrete ought to do it.
There's gold in these ideas, I tells ya.
> we all know that the primary use of balloons as
> applied to misbehaving prisoners
I thought the primary use of ballons as "applied" to prisoners involved an orifice insertion process, and that's all I'm going to say on the matter
Since when did the guys who run jails start using their imaginations? Maybe they need guys like me to start helping them out, coming up with wild and wacky ideas that would get prisoners roaming free again.
Or maybe they don't.
I just had a vision of releasing prisoners early, and using this balloon/camera thingy to track their every move. It hovers over them all the time, and feeds pictures back to some central point so their whereabouts can be monitored at all times. They're free to go wherever the rest of us can go, but they have this camera hovering over them all the time until their sentence is up. Think of the savings in jail accomodation!
Then the fatal flaw hit me: the ex-con goes for a job interview, holding a 1m balloon with a camera suspended below it, on a piece of string, in an office trying to describe how he'd be a great employee.
The local bad guys' public bar would look like a fairground, full of shiny balloons. "Mum, can we go play in that new inside park?" would be the cry from the kindergarten set.
Or imagine a typical NBA game. With the number of balloons that would be floating over the players, nobody would be able to watch the game. Hold on - there's no reason the balloons couldn't contain advertising.
Well, actually that's several fatal flaws, but I still think it has "weird and cool" merits that override the "it's a really, really dumb idea" issues. This idea has a really great application somewhere, but I just can't see it at the moment.
Gotta get more sleep tonight
OK, I will
Phone rings: "Let this be a warning to you: in Soviet Russia, spam *recipient* drives you crazy"
Hang up
Phone rings...
Putting together the responses to date, it seems like this is the story:
SVG: lots of potential; the ideal solution if the plugin is installed or easily installable; XML based
Flash: the pragmatic choice; installed nearly everywhere; works reliably on all platforms; can be driven via XML
VML: good fit for IE 5+ browsers, since it's supported in a default install; unsupported in Mozilla; deprecated technology; XML based
Java applet: slow to startup; JFreeChart looks very powerful provided startup times are acceptable and browser support is available
In case you can't tell, I'm facing the same problem as the original poster. I also can't deploy server-side solutions, so PHP, Perl GD, etc are all out.
Based on this, I'm inclined to build a generic XML solution, then rely on plugin detection and XSLT to deliver either SVG or VML as appropriate. That covers me for IE 5.5+ and anything with SVG support. The largest group that I don't have covered is Mac users and people with older browsers, and I could probably come up with a different XSLT to cope with them if the demand made it worth doing.
When you say OSS software isn't quite ready, and from the context I'm assuming you're talking about MS Office replacements and similar end-user-facing stuff, you have to remember that governments aren't full of people creating complex spreadsheets and Word documents.
Many/most government employees are "process workers"; people who use a very small number of programs (e.g. a Web browser) to perform largely repetitive tasks. There's very little knowledge or IT training these people need to do their jobs; what they need to know is (a) how to login, (b) how to start their application, (c) how to navigate through the screens and enter in data. They *don't* need training on KDE, Gnome, Unix file system layout etc.
For these people, I'd say OSS is well and truly ready, and has been for some time.
On that basis, modchips can't be illegal per se in the US - they're only illegal if they're used (as you say) to do things like play pirated games or develop emulation to let non XBox hardware play XBox games.
How, then, did MS manage to get that mod chip vendor (name escapes me...) to close down a few months ago? Sure MS is a big bully, but the vendor was in Asia somewhere so the DMCA wouldn't apply. I suppose MS could say "we'll take you to court till you're broke", but when there's no question of any valid law being broken (either in the vendor's country or the US) surely the case would be dismissed as frivolous, and MS would have to pay costs *and* reparations to the mod chip vendor (WHAT'S THEIR DAMN NAME??).
Am I missing something? If I was the chip vendor in that situation, I can't see why I wouldn't take MS on in court; I'd put up with a lot of legal pain and the likely loss of my business if I thought there was a good chance I'd be able to make a large chunk of loot out of reparations at the end of it all. Furthermore, finding a legal firm to bankroll me on a "no win, no fee" basis might be pretty easy if I could offer them e.g. 50% of the final payout.
> In any case there is a specific clause in the DMCA
> that allows for reverse engineering for the
> purpose of interoperability.
Warning: non-US citizen asking...
Based on this, could you then buy and mod an XBox, connecting it to your sound system as a media player system? If you have all your e.g. OGG files stored on a regular PC that runs as a file server (as I do), and play them from the modded XBox attached to your sound system, then you would have used the modded XBox solely for interoperability purposes.
I'm on the point of building exactly this setup in Australia now; would it be legal in the US on the basis of this interoperability clause?
They'll have lots of legacy Windows apps that will need to be progressively ported to Linux, so it makes sense to do this in a planned, application-by-application manner rather than go for a 'big bang' approach.
Furthermore, maybe only 50% of their users will need to run Windows/VMWare at any one time, so they may only need to buy half as many Windows licences as they have machines. Over time, this would decrease, as more and more apps get moved off Windows and onto Linux, and the number of users that need Windows decreases. And before anyone says "this is illegal", remember that we're talking about Germany, and they have much stronger laws restricting licence tying than they do in the US. I'm not an expert in German law, but this is quite likely to be "more legal" (if that term makes sense) in Germany than it would be in the US.
> Yeah, SVG is technically a W3C standard, but it's
> hardly a standard in actual web development.
Think about what ubiquitous SVG would do for browser based apps. You could do things like have graphs being updated from real-time data sources, with all sorts of interactivity possibilities. This *alone* is a massive step over what passes today for interactive data presentation and analysis capabilities, and has very big implications for all sorts of scientific and financial applications.
I'm aware you can do some of this using PNG, GIF or JPG pictures, but you don't get anything like the interactive capabilities that SVG makes possible using these formats.
I vote for adding SVG support to all browsers
*Finally*, I can start saying SVG is going to be supported natively in a browser, and pushing through projects on that basis.
..."
Until now, I've had to say you can use IE, then get an addon from Adobe. "What? Why doesn't MS support this SVG thing natively? What if Adobe decides to drop support for SVG; then what happens?
This is the best news I've read on Slashdot for a while
Does anyone know whether the govt guys that were involved in making the original agreement are still around? If so, are they free from legal action, or can they simply wipe the slate clean and get away with it?
If it's a totally different bunch of guys running the govt now, then they could conceivably justify it as being a dumb decision to enter into the agreement in the first place (using whatever logic they care to present...). If not, then maybe there's some negligence issues that can't be simply signed away with the swipe of a pen.
For those dollars, you aren't in the target market for a .NET solution.
.NET solution; you just need to enlighten them that it'll cost more than they've currently given you to build it, so they now have to choose between spending more or having a non-.NET solution.
You're not going to have much change out of $20k by the time you buy your server licences and hardware. You've either got to get more loot, or look for another solution (e.g. OSS-based). It's fine if your customer wants to have a
End of story
I won't have to fly over to the US and stand in that stupid "foreign nationals" line at LA airport any more, in order to get my vote recognised.
I can just hack in from here, from the comfort of my terrorist hideout in (insert "bad guy country" of the month)
> At the risk of getting into a partisan flamewar,
> what happened to Florida is suspect for
> election tampering
Yeah, I've heard there was a Republican somewhere on Slashdot, but I assumed it was only a rumor...
I frequently find myself in the same situation, and have switched to using Python wherever possible for the Q&D solution.
I've found that, when I write Python code, I generally get the structure of the code pretty right, though I may wind up skimping on documentation, may code stuff inline that should be separated out into functions, and so on. The step of converting Q&D code to production quality using Python is a whole lot smaller than in e.g. C.
Even if the final version has to be in C/C++, then I find implementing a Q&D version in Python first allows me to write better C/C++ code later on. Again, I suspect it's because using Python enables me to get the basic structure of the code 95%+ right on a first attempt, and then it's largely a matter of translating Python to C/C++.
Whatever you do, don't try creating a Q&D version in C/C++ if you can possibly avoid it. C/C++ code tends to be degenerate into unreadable/unmanageable crap much quicker when you're using it for Q&D code; they're great languages for production quality code, but just not that well suited to Q&D code.
IMHO, YMMV, etc...
I've worked in many government departments, always as an independent contractor. My experience is that government databases, and the apps that talk to these databases, are generally a mess.
I think there's a few issues involved:
- a lot of the code is written by people with skill levels that wouldn't be accepted in the outside world. I've got no doubt that there's highly skilful coders working for government, but there's a lot of duds as well and often your code is only as good as the weakest coder in the team. Furthermore, a lot of the weakest coders are the designated "experts" for legacy infrastructure, such as databases, and you find yourself having to rely on their input far more than you'd like
- a lot of government stuff has been outsourced, and generally the outsource partners seem to be less diligent for government contracts
- governments are forced to send everything out for tender, and may need to give work to the lowest bidder or face internal inquiries. In the private sector, you get to eliminate a lot of tender responses, either because you know the respondent is incompetent from past experience or the price is so low that the respondent couldn't possibly deliver a quality service; in government, you don't have the same luxury of filtering out the crap
- in the private sector, a lot of mainframers moved across to Unix, then Windows, in response to changing demand. While these people may not be expert in all 3 areas, they at least know enough to be able to hold a technical discussion with someone else from another area. In government, a lot of mainframers have been doing the same job forever, and you need to tie everything back to their narrow view of the world to get anything out of them whatsoever. You'd better not try to talk "Web Services" with these guys... There's a level of financial support and job security for these types of people in government that allows them to keep doing this, that simply doesn't exist in private enterprise
Now that I've pissed off everyone working in govt IT, I have to reiterate that some of them are extremely good. The issue is that the system doesn't work on a "survival of the fittest" system such as applies in private enterprise.
> ...objects written in one language are easily used
.NET with Java in this context; you need to compare it with the JRE.
.NET
> in every other language, so you can have 1
> developer using Perl, another using C#. Try that
> in Java. Try any cross language development in
> Java
You can't compare
I can and have written Python code that runs in the JRE, using Jython (a 100% Pure Java implementation of Python). It lets me mix and match Python and Java code, accessing objects across languages, and leverage the best of each language. I know you can run a LOT of other languages under the JRE; the number of available languages is at least comparable with those supported under
> It encourages good architectural practices.
.NET I haven't used to any great extent. I have used ASP a fair bit though
Disclaimer: ASP.NET is the one piece of
Serious question: How does ASP.NET encourage good architectural practices?
As I understand it, ASP.NET still suffers from the fundamental architecture problem of ASP in that it encourages you to mix presentation and business logic. This is a nightmare for design and ongoing maintenance.
Is this correct, or have I missed something?