Anyone can write malicious code in C, but it takes a real pro with real knowledge to even begin to try that with Java.
I assume you're joking, since Java is as vulnerable to typical attacks like SQL injection just as much as many other languages, with just the same sort of careless slip in a couple of lines by a tired programmer...
No, but when you have 1/3 of the world economy, you become a hole continent.
Hmm... I'm sure there must be some measure where the US economy represents 1/3 of the world total. Some measure of overall deficit, perhaps? Annual pollution emissions?
I guess that external reputation doesn't botter US anymore. There is not much what the rest of the world can do to harm them
On the contrary, the ever more aggressive stance of US corporations, and the apparently carefree support the US government so willingly gives them, is increasingly loading the dice against competitors in other countries, and hence the economies of those countries. Combined with a lack of genuine willingness to compromise, and it's not hard to see other major economic powers like Europe or major growing economies like China doing a lot more to cut the US out (c.f. Microsoft vs. the EU, for example, and now consider similar treatment being given routinely to all major US corporations regarded as a threat to the local economy in the places I mentioned).
Consider this: here in the UK, I buy a lot of expensive stuff that says "made in Japan", a lot of everyday stuff that says "produce of somewhere in Europe", and pretty much nothing that says "made in the US". In other words, it's not consumer goods driving international trade with the US, it's business transactions. And the businesses are starting to get upset.
Bottom line: whatever the US economic powers may collectively believe, the rest of the world could do without the US a lot more than the US could do without the rest of the world. In the end, when the US economic bubble bursts, as it surely will before much longer, the biggest thing that's going to halt the resulting nosedive is foreign support -- if there's any left.
Well, if you are American you don't have to fear being killed - yet. If you are an Iraqi or an Afghan however you can expect to be tortured and killed, or perhaps locked up in some god forsaken hole in Cuba for years on end.
Also <inserts list of numerous other countries, including several "allies", whose citizens were/are held without due process at Gitmo>.
I think one of the saddest things about that whole sorry issue is the string of US court rulings that the guys being held without trial etc. weren't subject to the protections of the US constitution (because they weren't held on US soil and/or weren't US citizens) or the Geneva Convention (because they're not classed as prisoners of war). The US authorities have carved out a little black box and put them all in there to conveniently side-step any responsibility for their actions. It's one step removed from simply "disappearing" them, and the step is a very short one.
If the US cannot demonstrate the rule of law by example then it does not deserve anymore respect than a warlord in a cave.
Don't worry. In reports like this, it isn't getting any more respect than a wardlord in a cave, either.
It's funny how governments like to rely on reports from independent groups like Amnesty to justify, say, invading a country and removing an evil dictator. Then the same groups become "unreliable" or "ill-informed" or something when you look at their reports from those same countries after the invasion, or their reports on the behaviour of governments closer to home.
Sometimes I wonder whether US Gov Corp. realises how much it's damaged the reputation of the US internationally with things like Gitmo. Other times I wonder whether they do know, and just don't care. Either way, it's too bad; in the long run, no man (or nation) is an island.
There simply wasnt any need for that particular software. If a software is needed by a large number of people then it will get forked succesfully and will grow.
But if there's a large number of people wanting it, chances are a commercial development wouldn't cease either. What counts is when a small number of users want something to continue development, and in that case, I'm not convinced OSS has yet demonstrated any compelling advantage over CSS.
Your entire argument assumes that people cant become familiar with a large piece of code if they didnt develope it. I have to say that isnt true. If the person is a decent coder s/he can learn a large project such as OOo within a year, give or take depending on previous involvement.
Can you cite any studies to back that up? IME, having joined several large, established projects when starting new jobs myself, it takes at least 1-2 years for a competent developer to get up to speed on a medium-sized project. That, of course, is working on it full-time, with training and support from existing developers, and with access to all the docs etc. Since OOo is a very large project, I don't see how a team of your "decent coders" is going to find their way around and take over the development at all effectively within a year.
Support for those things dont exist because the need doesnt exist
If your project has any future, then the need exists. IME, even most professionals don't write as much documentation as they should. Their answer is usually "we don't need to", but then they're surprised when subtle bugs appear a year or two later, and the new guys take ages to get up to speed and constantly seem to be asking time-consuming questions.
Oh, I don't think so. Per-seat licensing of Windows, Office Professional, and whatever other tools are recurring costs at each upgrade.
As are the time costs of installing new versions of Linux and its tools. I'm still using the same versions of Windows and Office as when I started this job three years ago, and we've upgraded Visual Studio once in that time. How many times do you think we'd have had to upgrade our Linux-based tools during the same period?
But once you begin deploying in the hundreds or thousands it really adds up.
As does the time required to configure Linux and its brethren for all those extra staff. And the financial cost of the Windows platform upgrades is still less than a day of employee time, no matter how many employees you're doing it for.
But as anti-big brother as I am I think this is perfectly
reasonable. While you're at work they own your ass--and they own the computer and
they own the network. They have the right to do whatever they want with their
property.
In isolation, that may be true for you, though here it's not; there are all
kinds of thing my employer doesn't own even when I'm at work. In any case,
whether it's a good thing for anyone is a different question.
If my boss is reading this -- which I'm writing during my lunch hour, but
using company property -- and finds it objectionable, he's welcome to tell me to
stop, and I will do so. (And then I'll give them my notice, along with probably
half my colleagues.) Anyone who's monitoring my PC's Internet traffic would
probably think I was spending half my day on the web too, since I tend to take a
break for a couple of minutes every half hour or so, and often surf to a
favourite site or two for a few seconds during that time. Then again, I also
record an extra few minutes a day as breaks on my time sheet, so I consider them my time and
have no guilt about doing this.
Here in the UK, there has always been a certain level of understanding that
employees aren't robots, they're human beings. Hence a small amount of personal
phone calls is considered reasonable. (As an aside, IIRC there are legal limits
here on the monitoring of those calls by employers as well, even though they're
done using company phones.) This culture is best for everyone: staff don't get
stressed because they don't know their child got home from school OK or they have
time to arrange meeting their SO after work, while the company gets more loyal
and productive workers as a result.
It would be better for all the same reasons if e-mail were simply treated the
same way, and except for the sort of legal idiocy the US in particular is
infamous for, it could be. Unfortunately, we seem to have wound up in an absurd
situation where anything sent by any employee from a company e-mail address can
cause liability for the company, whether authorised or not. I don't know how or
why that happened, and someone should really fix it as a priority, since even the
most diligent employer could get caught out here. I can't write to someone and
make a deal on behalf of my employer; I don't have the authority to do that,
whether I print the letter on company headed paper or not. Why e-mail should
carry any more weight, I don't know, and if it doesn't, then it should be made
clear that employers can't be held liable for it and therefore shouldn't have to
worry about it. At that point, the scanning issue is pretty much irrelevant.
ok lets take examples from history where the company stopped making the software: Netscape, AOL stopped netscape dev and handed it off to mofo now what has happend there? a robust and full growth.
Or an entirely new project based around Gecko that happened to use the same name, after several years of effective non-existence while the competition moved on, depending on your perspective.
Basically you are arguing against the strenght of OSS, forking;
I'm not arguing against it, I'm simply saying that it's not some silver bullet to fix all your future-proofing problems. Since there have been very few really successful forks, particularly when the original project has basically shut down, I don't think that's an unreasonable position. If and when there's more comprehensive evidence of successful forks on more "average" OSS projects, I'll consider the approach a proven one that might be worth risking your business for, but not until then.
even if Sun stops making OOo tommorow guess what you are gonna have the community come forward and take over.
Do you realise that almost all of the development on that project is actually done by Sun staff at present? Where are all these knowledgeable and skilled "community" developers going to come from, if there are only a handful of them even now when Sun is supporting the project?
Now your comments about docs, the docs are created in a OSS project if the user needs'em for example take a look at great enduser OSS: ubuntu, firefox, etc. they all have greate docs.
And they are among the big success stories of OSS, with exceptionally large support. Unfortunately, most OSS projects aren't like that, and the same rules don't necessarily apply.
Sure the game was a financial success but lets face it, the entire Matrix series never fully recovered from the shock.
In fairness, I'm not sure the Matrix series ever fully recovered from Reloaded and Revolutions. The original was one of the best SF movies ever made. The sequel was a mediocre action movie. The third was an insult to the whole concept.
That's not necessarily true at all. If you're writing CGI code in C++ (and not so long ago, a heavy majority of the most-visited web sites in the world did; I don't know whether they still do) then you have to talk to your back-end systems somehow. On a Microsoft platform, that probably means some hideous code written using at least one of Win32 and MFC.
With in house Linux (or UNIX) personnel and a large deployment - of course you'll save big.
I've heard this argument a lot, and yet... At the hourly cost of employing most guys at the office where I work (mostly developers, and a few tech support/sales guys), the amount my employer pays for a Windows licence is worth a little over an hour. Office is a couple more, and Visual Studio a couple more. Since there's a very good chance that someone in the office will know how to do just about anything with those key products, ongoing support costs are close to nil.
In other words, even if we were all instantly and permanently as productive with Linux and a Linux-based office suite and development tool, and they also required no ongoing support costs because we could pretty much do it all ourselves, the most switching to Linux would save my employer is the cost of a day of my time. It would take that day just to set the new system up.
Now, there are many UNIX fans at the office where I work, and we all know at least the basics since we develop for several UNIX-ish platforms, including Linux. Even so, I don't for an instant believe we'd be as productive using the Linux-based tools (which we're free to do if we want, but almost no-one does, even the Linux fans).
In summary: this is an office full of technically competent people with no particular love of Microsoft and a certain fondness for Linux, and yet I can't make a genuine case for switching on cost saving grounds. Unless I'm missing some major financial consideration, that pretty much kills the argument for other offices without either the technical expertise or the Linux fans dead. So where are the big savings I keep being told about coming from?
The FUD that Microsoft spreads, and that people like you seem determined to help them spread, goes like this: the ancillary costs associated with Linux are higher than those associated with Windows; higher enough, in fact, that Windows is cheaper overall. Which, now that I think of it, doesn't even deserve a fancy name like "FUD" -- the simple, old-fashioned word "bullshit" is quite sufficient.
If (big "if") the data in the article is representative, then it would appear that you are mistaken.
In fact, if you scan some of the articles/comments that have been in the Apple section recently, it seems pretty clear that one of Apple's major reasons for backing Intel is the DRM. They're going for domination of the multimedia field, and presumably the big media corps wouldn't play without it.
Seems kinda sad; you'd think if anyone was smart enough to realise that DRM was a losing bet and going for legitimate+user friendly was a better one, it would be Apple. Maybe next time...
A good developer should be able to learn a new system quickly enough to program for it.
A really top notch developer with generally wide experience will pick up a new OS/library combination like this well enough to do the basics in a day or two, and then spend the next several weeks learning the subtleties and idioms well enough to actually write good code on that platform.
An average developer is around an order of magnitude less productive than the guys at top end of the scale.
At this point, either calling in those consultants or switching to a platform your guys already know well sounds like a better option in the real world.
Free Software mostly eliminates those problems, because there's always a new vendor that can take over the management of your current technology.
That's really not true, you know. If Sun stopped supporting Star/OpenOffice, or the guys at MySQL gave up and went home, I'd give you great odds that it would pretty much kill future development of the corresponding product as well, open source or not. You might get the occasional bug fix or minor patch, but that's probably it.
The harsh reality is that just because you've got the source code to something doesn't mean you can instantly be an expert in how it's been designed and all the little things that were learned along the way, nor competent to continue development as if you were the original dev team. This has nothing to do with being open source; classical closed source businesses have been facing this dilemma as a matter of HR management for as long as there's been software development. Moreover, even a relatively weak dev team doing this stuff professionally usually generates better documentation than most F/OSS projects seem to, and there's usually more continuity even if some of the original development staff leave.
Open source projects that are widely distributed in more than one version by somewhat independent groups -- in other words, Linux -- are a good bet for future-proofing. Open source projects that are widely distributed in only one version with only a few core developers -- in other words, pretty much everything else, including the other big names -- are not good bets. In fact, if future-proofing is your biggest concern, the most robust option is to get it written in-house, with an emphasis on good documentation and testing procedures, and with solid management overseeing the work. Failing that, you might well be better off going with something provided by a megacorp that employs vast numbers of developers and has vast financial reserves.
And what do they do after you register? Can your site be shut down if they don't "agree" with your site (ideas)?
To put it bluntly: yes. And if your ideas are controversial and you refuse, the sanctions will probably be severe. Don't forget that you're talking about China, where human rights is still an almost mythological term, and when dissidents are shot the invoice for the bullet goes to their family.
I suspect you'll be proved right. I've played many games with flashy graphics and sound effects and given up halfway through. The best games, invariably, are those that require real strategy and/or provide real plot development.
Graphics and sound effects can make or break a game with that kind of gameplay, but nothing can make a game without it.
The title says it all, really. If the technical information underlying the technical evidence and affirming its validity cannot be supplied, the charges are dismissed. Closed Source -> Charges Dismissed.
The problem is that those two statements are not equivalent.
I think we're in strong agreement. If evidence from a system is to be used, that system must (at minimum) be properly reviewed and approved by a competent and independent authority. However, once that review has taken place, requiring the source code to be opened up routinely for scrutiny may not be necessary if there's nothing else unusual about the case. Otherwise you'll get every speeding ticket case lasting months while every defence conducts the same "expert review" of the same thousands of lines of source code, which won't do anything to promote justice and fairness within the legal system.
That's because all four Seminole County criminal judges now use the same standard: If a DUI defendant asks for a key piece of information about how the machine works -- its software source code -- and the state can't provide it, the breath test is rejected.
It's tough to say whether that's entirely fair; on the one hand, source code can be checked, but on the other, how do you know that's what was actually running on the machine at the time anyway? What really matters is that the algorithm/process is reliable, rather than the actual source code. If that algorithm/process has been properly reviewed and approved, whinging about lack of source code sounds more likely to let people off on a technicality than something actually in the interests of justice. There comes a point where you have to trust to review and approval processes, because you can't individually review every line of code in detail in every case.
However, one way or another, those submitting the technical evidence should surely be required to provide detailed information about how it works for a defence expert witnesses to review, or accept that their technical evidence may not carry full weight with a judge or jury otherwise. If, as was suggested in this article, the method had actually been modified since the original approval was granted, then that approval should carry no weight here, and if that undermines the technical evidence as well and nothing else (e.g., the source code) is presented to support the validity of the changes, then that evidence carries no real weight either.
Slashdot Mirror
I assume you're joking, since Java is as vulnerable to typical attacks like SQL injection just as much as many other languages, with just the same sort of careless slip in a couple of lines by a tired programmer...
Hmm... I'm sure there must be some measure where the US economy represents 1/3 of the world total. Some measure of overall deficit, perhaps? Annual pollution emissions?
On the contrary, the ever more aggressive stance of US corporations, and the apparently carefree support the US government so willingly gives them, is increasingly loading the dice against competitors in other countries, and hence the economies of those countries. Combined with a lack of genuine willingness to compromise, and it's not hard to see other major economic powers like Europe or major growing economies like China doing a lot more to cut the US out (c.f. Microsoft vs. the EU, for example, and now consider similar treatment being given routinely to all major US corporations regarded as a threat to the local economy in the places I mentioned).
Consider this: here in the UK, I buy a lot of expensive stuff that says "made in Japan", a lot of everyday stuff that says "produce of somewhere in Europe", and pretty much nothing that says "made in the US". In other words, it's not consumer goods driving international trade with the US, it's business transactions. And the businesses are starting to get upset.
Bottom line: whatever the US economic powers may collectively believe, the rest of the world could do without the US a lot more than the US could do without the rest of the world. In the end, when the US economic bubble bursts, as it surely will before much longer, the biggest thing that's going to halt the resulting nosedive is foreign support -- if there's any left.
Also <inserts list of numerous other countries, including several "allies", whose citizens were/are held without due process at Gitmo>.
I think one of the saddest things about that whole sorry issue is the string of US court rulings that the guys being held without trial etc. weren't subject to the protections of the US constitution (because they weren't held on US soil and/or weren't US citizens) or the Geneva Convention (because they're not classed as prisoners of war). The US authorities have carved out a little black box and put them all in there to conveniently side-step any responsibility for their actions. It's one step removed from simply "disappearing" them, and the step is a very short one.
Don't worry. In reports like this, it isn't getting any more respect than a wardlord in a cave, either.
It's funny how governments like to rely on reports from independent groups like Amnesty to justify, say, invading a country and removing an evil dictator. Then the same groups become "unreliable" or "ill-informed" or something when you look at their reports from those same countries after the invasion, or their reports on the behaviour of governments closer to home.
Sometimes I wonder whether US Gov Corp. realises how much it's damaged the reputation of the US internationally with things like Gitmo. Other times I wonder whether they do know, and just don't care. Either way, it's too bad; in the long run, no man (or nation) is an island.
But if there's a large number of people wanting it, chances are a commercial development wouldn't cease either. What counts is when a small number of users want something to continue development, and in that case, I'm not convinced OSS has yet demonstrated any compelling advantage over CSS.
Can you cite any studies to back that up? IME, having joined several large, established projects when starting new jobs myself, it takes at least 1-2 years for a competent developer to get up to speed on a medium-sized project. That, of course, is working on it full-time, with training and support from existing developers, and with access to all the docs etc. Since OOo is a very large project, I don't see how a team of your "decent coders" is going to find their way around and take over the development at all effectively within a year.
If your project has any future, then the need exists. IME, even most professionals don't write as much documentation as they should. Their answer is usually "we don't need to", but then they're surprised when subtle bugs appear a year or two later, and the new guys take ages to get up to speed and constantly seem to be asking time-consuming questions.
As are the time costs of installing new versions of Linux and its tools. I'm still using the same versions of Windows and Office as when I started this job three years ago, and we've upgraded Visual Studio once in that time. How many times do you think we'd have had to upgrade our Linux-based tools during the same period?
As does the time required to configure Linux and its brethren for all those extra staff. And the financial cost of the Windows platform upgrades is still less than a day of employee time, no matter how many employees you're doing it for.
In isolation, that may be true for you, though here it's not; there are all kinds of thing my employer doesn't own even when I'm at work. In any case, whether it's a good thing for anyone is a different question.
If my boss is reading this -- which I'm writing during my lunch hour, but using company property -- and finds it objectionable, he's welcome to tell me to stop, and I will do so. (And then I'll give them my notice, along with probably half my colleagues.) Anyone who's monitoring my PC's Internet traffic would probably think I was spending half my day on the web too, since I tend to take a break for a couple of minutes every half hour or so, and often surf to a favourite site or two for a few seconds during that time. Then again, I also record an extra few minutes a day as breaks on my time sheet, so I consider them my time and have no guilt about doing this.
Here in the UK, there has always been a certain level of understanding that employees aren't robots, they're human beings. Hence a small amount of personal phone calls is considered reasonable. (As an aside, IIRC there are legal limits here on the monitoring of those calls by employers as well, even though they're done using company phones.) This culture is best for everyone: staff don't get stressed because they don't know their child got home from school OK or they have time to arrange meeting their SO after work, while the company gets more loyal and productive workers as a result.
It would be better for all the same reasons if e-mail were simply treated the same way, and except for the sort of legal idiocy the US in particular is infamous for, it could be. Unfortunately, we seem to have wound up in an absurd situation where anything sent by any employee from a company e-mail address can cause liability for the company, whether authorised or not. I don't know how or why that happened, and someone should really fix it as a priority, since even the most diligent employer could get caught out here. I can't write to someone and make a deal on behalf of my employer; I don't have the authority to do that, whether I print the letter on company headed paper or not. Why e-mail should carry any more weight, I don't know, and if it doesn't, then it should be made clear that employers can't be held liable for it and therefore shouldn't have to worry about it. At that point, the scanning issue is pretty much irrelevant.
I think the line
kinda gave that away already.
To be more precise, if the development staff need crunch time to finish a software product, then chances are that the management did something wrong.
Or an entirely new project based around Gecko that happened to use the same name, after several years of effective non-existence while the competition moved on, depending on your perspective.
I'm not arguing against it, I'm simply saying that it's not some silver bullet to fix all your future-proofing problems. Since there have been very few really successful forks, particularly when the original project has basically shut down, I don't think that's an unreasonable position. If and when there's more comprehensive evidence of successful forks on more "average" OSS projects, I'll consider the approach a proven one that might be worth risking your business for, but not until then.
Do you realise that almost all of the development on that project is actually done by Sun staff at present? Where are all these knowledgeable and skilled "community" developers going to come from, if there are only a handful of them even now when Sun is supporting the project?
And they are among the big success stories of OSS, with exceptionally large support. Unfortunately, most OSS projects aren't like that, and the same rules don't necessarily apply.
In fairness, I'm not sure the Matrix series ever fully recovered from Reloaded and Revolutions. The original was one of the best SF movies ever made. The sequel was a mediocre action movie. The third was an insult to the whole concept.
That's not necessarily true at all. If you're writing CGI code in C++ (and not so long ago, a heavy majority of the most-visited web sites in the world did; I don't know whether they still do) then you have to talk to your back-end systems somehow. On a Microsoft platform, that probably means some hideous code written using at least one of Win32 and MFC.
I've heard this argument a lot, and yet... At the hourly cost of employing most guys at the office where I work (mostly developers, and a few tech support/sales guys), the amount my employer pays for a Windows licence is worth a little over an hour. Office is a couple more, and Visual Studio a couple more. Since there's a very good chance that someone in the office will know how to do just about anything with those key products, ongoing support costs are close to nil.
In other words, even if we were all instantly and permanently as productive with Linux and a Linux-based office suite and development tool, and they also required no ongoing support costs because we could pretty much do it all ourselves, the most switching to Linux would save my employer is the cost of a day of my time. It would take that day just to set the new system up.
Now, there are many UNIX fans at the office where I work, and we all know at least the basics since we develop for several UNIX-ish platforms, including Linux. Even so, I don't for an instant believe we'd be as productive using the Linux-based tools (which we're free to do if we want, but almost no-one does, even the Linux fans).
In summary: this is an office full of technically competent people with no particular love of Microsoft and a certain fondness for Linux, and yet I can't make a genuine case for switching on cost saving grounds. Unless I'm missing some major financial consideration, that pretty much kills the argument for other offices without either the technical expertise or the Linux fans dead. So where are the big savings I keep being told about coming from?
If (big "if") the data in the article is representative, then it would appear that you are mistaken.
In fact, if you scan some of the articles/comments that have been in the Apple section recently, it seems pretty clear that one of Apple's major reasons for backing Intel is the DRM. They're going for domination of the multimedia field, and presumably the big media corps wouldn't play without it.
Seems kinda sad; you'd think if anyone was smart enough to realise that DRM was a losing bet and going for legitimate+user friendly was a better one, it would be Apple. Maybe next time...
A really top notch developer with generally wide experience will pick up a new OS/library combination like this well enough to do the basics in a day or two, and then spend the next several weeks learning the subtleties and idioms well enough to actually write good code on that platform.
An average developer is around an order of magnitude less productive than the guys at top end of the scale.
At this point, either calling in those consultants or switching to a platform your guys already know well sounds like a better option in the real world.
Win32 and/or MFC.
That's really not true, you know. If Sun stopped supporting Star/OpenOffice, or the guys at MySQL gave up and went home, I'd give you great odds that it would pretty much kill future development of the corresponding product as well, open source or not. You might get the occasional bug fix or minor patch, but that's probably it.
The harsh reality is that just because you've got the source code to something doesn't mean you can instantly be an expert in how it's been designed and all the little things that were learned along the way, nor competent to continue development as if you were the original dev team. This has nothing to do with being open source; classical closed source businesses have been facing this dilemma as a matter of HR management for as long as there's been software development. Moreover, even a relatively weak dev team doing this stuff professionally usually generates better documentation than most F/OSS projects seem to, and there's usually more continuity even if some of the original development staff leave.
Open source projects that are widely distributed in more than one version by somewhat independent groups -- in other words, Linux -- are a good bet for future-proofing. Open source projects that are widely distributed in only one version with only a few core developers -- in other words, pretty much everything else, including the other big names -- are not good bets. In fact, if future-proofing is your biggest concern, the most robust option is to get it written in-house, with an emphasis on good documentation and testing procedures, and with solid management overseeing the work. Failing that, you might well be better off going with something provided by a megacorp that employs vast numbers of developers and has vast financial reserves.
To put it bluntly: yes. And if your ideas are controversial and you refuse, the sanctions will probably be severe. Don't forget that you're talking about China, where human rights is still an almost mythological term, and when dissidents are shot the invoice for the bullet goes to their family.
Sorry.
Dear Senior Job Application Consumer...
I suspect you'll be proved right. I've played many games with flashy graphics and sound effects and given up halfway through. The best games, invariably, are those that require real strategy and/or provide real plot development.
Graphics and sound effects can make or break a game with that kind of gameplay, but nothing can make a game without it.
It's the turnout of 105% that gives it away. :-)
The problem is that those two statements are not equivalent.
I think we're in strong agreement. If evidence from a system is to be used, that system must (at minimum) be properly reviewed and approved by a competent and independent authority. However, once that review has taken place, requiring the source code to be opened up routinely for scrutiny may not be necessary if there's nothing else unusual about the case. Otherwise you'll get every speeding ticket case lasting months while every defence conducts the same "expert review" of the same thousands of lines of source code, which won't do anything to promote justice and fairness within the legal system.
It's tough to say whether that's entirely fair; on the one hand, source code can be checked, but on the other, how do you know that's what was actually running on the machine at the time anyway? What really matters is that the algorithm/process is reliable, rather than the actual source code. If that algorithm/process has been properly reviewed and approved, whinging about lack of source code sounds more likely to let people off on a technicality than something actually in the interests of justice. There comes a point where you have to trust to review and approval processes, because you can't individually review every line of code in detail in every case.
However, one way or another, those submitting the technical evidence should surely be required to provide detailed information about how it works for a defence expert witnesses to review, or accept that their technical evidence may not carry full weight with a judge or jury otherwise. If, as was suggested in this article, the method had actually been modified since the original approval was granted, then that approval should carry no weight here, and if that undermines the technical evidence as well and nothing else (e.g., the source code) is presented to support the validity of the changes, then that evidence carries no real weight either.