The ideal way of doing things is to write as much as you can in the scripting language. This is almost always faster and more efficient in terms of programmer time. Then, you go back and redo the speed critical bits and pieces in C or C++.
I've heard that a lot, and certainly the argument has merit.
Where it sometimes breaks down, IMHO, is that learning a new language isn't free. It's a myth that any good programmer can learn a new language in a week. Sure, they can learn the basic syntax, and if they're familiar with the particular paradigms they're using (by which I mean OOP, functional programming, whatever) then they'll be able to apply those principles fairly readily. But there's a world of difference between that and the kind of clean, idiomatic, easily maintainable code than a good programmer with a lot of experience in the specific language(s) he's using would write.
I predict that this is going to be one of the bigger factors holding back $SCRIPTING_LANGUAGE from wider usage for a long time. There are simply too many almost isomorphic scripting languages with a significant, but still small, user base, and while they offer similar advantages, you can't just switch a whole dev team from one to another for the reason above.
I think this is also a big reason C++ remains popular, particularly with stronger programmers who make the effort to learn its idioms, who for some probably related reason rarely seem to encounter all these dreadful weaknesses C++ is supposed to have... In a nutshell, C++ is a single language with sufficient tools to work effectively with both low-level details and high-level designs, and perhaps a strong programmer with such a tool will be more effective than a programmer with two more suitable tools, but only moderate skill with either.
A fairer question might be what's right about it.;-)
Seriously, the string-handling in standard C++ suffers horribly from design-by-committee.
To give the most obvious example, there's no way to write a literal std::string, which in turn means that basic syntax like "a"+"b" doesn't work, even though the + operator is used to concatenate std::string objects.
There are also glaring inconsistencies in usage, because of all the history using char*, and then numerous class libraries inventing their own string type. Take a look at the I/O streams interfaces in C++: you can't even pass a std::string as a filename without converting to a C-style string first.
Then there's the interface, which is a bit Jack-of-all-trades, and bloated with it.
And of course, even with a real string type now -- surely a prerequisite for claiming any meaningful string support at all! -- C++ still doesn't provide standard support for basic string processing using regular expressions. Compared to a more text-focussed language like Perl, well, there's no comparison.
Fortunately, a number of these sillies are in line to be fixed in the next revision of C++, but for now, the language's string support is pretty much a joke.
He was president. He received a blaring PDB warning (by PDB standards).
Just out of interest, how many such "glaring warnings" do you think a national commander-in-chief receives during his career? How many of them do you think appear so glaring before they become reality?
While no man can do everything, the person entrusted to be the president of the United States sure the hell can come back from vacation when he receives credible warnings of imminent terrorist attacks on U.S. soil.
Do you really think POTUS can't give any orders he feels necessary immediately, even when he's on vacation?
If you think he misjudged the briefing, fine. If you think he did nothing when he should have acted, fine. But "coming back from vacation" is just PR.
The primary cause of death through malnutrition is the deliberate action of dictatorial governments.
In many cases, yes it is. In some cases, it's not, and the work of aid agencies that rely primarily on charitable contributions to fund their work is critical. Just look at what's been achieved near the tsunami site, where the public collectively acknowledged the problem and voted with their wallets to do something about it. Yet more people die in Africa every year than died as a result of the tsunami, and we do relatively little to help.
The solution is to overthrow those governments and establish better ones. I doubt that such a solution would be widely tolerated.
We (several first world countries) have just overthrown the governments of both Afghanistan and Iraq for far less constructive purposes. Do you really think anyone at the top of, say, the US or UK governments is ever going to be held accountable? Hell, the US re-elected Bush despite that, and sadly it looks as though we in the UK are about to re-elect Blair (though probably with a much-reduced majority, not least because of his actions relating to Iraq).
Some of the stuff that goes on in Africa makes the Middle East dictatorships look pleasant, and there's an obvious humanitarian motivation for waging war. Of course, they don't produce as much oil, and presumably it would therefore be somehow more illegal under international law to topple people like Mugabe than it was to topple Hussein or the Taleban?
This memo was obviously written by someone trying to get Bush's attention to actually give a shit about going after terrorism.
And how many memos do you think the most powerful man in the world gets every day, trying to draw his attention to some concern that the author thinks is the most important thing in the world?
Some groups had been advocating a tsunami warning system for years, which would have saved orders of magnitude more lives than foreseeing 9/11, at a fraction of the cost of clearing up the damage the tsunami caused when it hit.
There are enough resources in the world to feed every human alive a good diet, yet many thousands die through malnutrition every year in third world countries, because the rest of the world doesn't do enough to help them.
If a polar ice cap melts next year, causing millions of deaths as coastal cities are flooded all over the world, the critics will be saying that the environmentalists had been warning of this for years.
If the super-volcanic eruption happens and blacks out the whole US, also killing millions, someone will point out that there was a documentary on TV a few months ago warning about the possibility.
If an unpredicted asteroid hits the earth tomorrow and wipes out half the population of the planet, someone will be saying that the warnings were there and billions should have been invested in a super-defence-network-satellite-system-thing.
I don't know whether Bush really missed something he should have seen, or indeed if the intelligence community collectively did. Maybe so, in which case that's a problem that needs fixing. But, while I'm no fan of GWB, I do have realistic expectations, and I recognise that there are limits to what one man -- however powerful -- can see and do.
Are you sure it wasn't just a made-up concept with a pseudo-funky name that the new designer invented to sound "cool"?
Half of that redesign wasn't a change in graphic design, it was a complete change in content and implied a complete change in process. Chances are the designer had absolutely no idea what processes this document actually fits into in reality, and has equally little idea whether the changes in content he made would offer any value, or even whether they'd actually make the document less useful.
We should have just gone to red alert, switched on the force fields, and tractor-beamed the planes into the docking bay, Captain.
he was twice forced to resign over allegations of sleaze?
Whatever you think of his politics, he's a very shrewd operator. He's long been a close friend of [UK Prime Minister] Tony Blair, and despite the bad press (justified or otherwise) that some of his actions have drawn, he's continued a high-flying political career. His resignations have avoided a lot of the amateurish fall-out that some of his peers (David Blunkett and Clare Short, for two examples) incurred by messing up the PR, by taking him out of the limelight just long enough for the public to forget any transgressions he might have committed. Of course, they also get him nicely spoofed on the political satire shows, but such is the price of celebrity.;-)
It frustrates me because I don't think older open source projects should just mysteriously break. For example, I have an older sublaptop which I got pretty cheaply on Ebay. There is some someware I would like to try on this machine, such as the last pre-Gecko version of Mozilla (which had a not of Netscape 4 code in it), that I can't get to compile in Gcc 3.4 because too much has changed since then.
I understand your frustration, but it sounds like the original code wasn't written in standard C++ when it could have been; the standard has been around since about '97 for practical purposes, after all. This is a great example of why we have standards: if you write code to the standard, it will compile on any standard-compliant compiler (yeah, I know), and it will continue to compile on any future standard-compliant versions. If you rely on non-standard features or implementation-defined or undefined behaviour, then you're trusting to fate whether your code will continue to compile and work the same way in future.
In other words, this is certainly a real problem, but I think you're attributing the blame to the wrong source. When you write
Or would be able to, except for the fact that the people who write GCC continue to insist on breaking older code.
you imply that the older code wasn't broken already, which probably isn't the case -- certainly if it was written as recently as 2002!
FWIW, I work with a lot of code that was written pre-standard, and the pre-standard compilers to match, every day, and it can be very frustrating when things that are trivial in standard C++ take extra code clutter and testing time to achieve "just because". I feel your pain!:-)
Not to disagree with your general post but you understand the whole Free Software thing makes it impossible for there to be technology and extensions in GCC that competitors can not adopt right?
So I've been told.
Unfortunately, in the real world, porting that technology isn't free: at the very least, it takes time and resources. And since those extensions aren't standard, it may well not be the best use of whatever limited time and resources other compiler writers have.
Moreover, the fact that GCC is GPL'd makes lifting its algorithms straight out a massive no-no for any commercial compiler vendor I can think of. Without wanting to restart the usual pointless holy war, this is an obvious case where the GPL being free-as-in-FSF and not free-as-in-completely may well be inhibiting reuse of existing code/design work.
This all sounds a bit harsh, so if you can name me any other C++ compiler that effectively implements all the C++ extensions used in the major Linux-related projects, I'll concede that the reuse isn't being completely inhibited.
Re:GCC 4.0's biggest winner is probably KDE
on
GCC 4.0.0 Released
·
· Score: 2, Funny
The slow part of KDE is the compile time...
Damn. You have to compile it? No wonder mine hasn't loaded yet...
I don't expect people to "pull a number out of their ass" when I ask for an estimate. However, I expect them to be able to formulate a list of requirements they have which, when addressed, will let them give me an accurate estimate.
That's fine, but what if the amount of time required to list those requirements accurately could be 80% of the total development time for the project? Anyone can give you an estimate for the release date five minutes before they release the code.;-)
Seriously, I see this all the time at the moment. I'm adding new features to a relatively old piece of software. My last project was timetabled to take about six weeks of development, but lasted nearly a year. The new development work I did actually took (give or take a couple of days) the amount of time I'd estimated, and met its spec after that time. The problem was that then they found a zillion bits of old code that didn't meet their spec, hadn't been tested sufficiently to discover this a decade ago, but now weren't up to the job of driving my nice new functionality. Fixing those extended the length of the project as a whole by an order of magnitude.
Now, I could have told the project managers that was going to happen. But until I'd spent the several months I took investigating the older code to identify the numerous special cases that would need removing in order to reduce it to a systematic, effectively testable design, I couldn't have told them how long it was going to take. Once you've got that knowledge, writing the code itself is barely significant in the amount of time it takes.
I've heard this argument..... that Word is so powerful..... but every time I come close to believing it, I find evidence of a document with about fifteen different fonts in it, formatted using spaces, and a manually-entered table of contents.
Sure. Random, untrained users often fail to take advantage of even the basic features of their software. As the saying goes, if you think training is expensive, try ignorance.
However, those who produce documents for a living and take an hour or two to learn the basics of word processing at some point in their lives will gain better results faster by using stylesheets, automatic mail merge, automatic cross-references and the like. Actually, those are three of Word's weaker features, but even then it remains some way ahead of OpenOfficer Writer, at least for now.
If MS Word could import and export.SXW files natively, there would be no need for MS Word in the first place. It's only the fact that Word -- and nothing else -- can read.DOC files properly, that keeps Microsoft selling it.
I was going to mod you down for trolling, but I'd rather set the record straight.
There is clearly more to new versions of Word than being able to open previous versions' documents. If there weren't, then far more businesses would have given up upgrading by now. The average home user may not use all (or perhaps even any) of the new features, particularly in the most recent incarnations where they've been heavily focussed on interoperability with other Office apps, but many business users do.
Even without the new features, Word remains a far more powerful tool than OpenOffice.org Writer (and I say this as someone who works with both apps regularly, on a wide range of document types). OO.o has enormous potential, and I'm grateful to the dev team for giving me such a useful product that I can install for free at home, but let's keep some perspective. I can't think of any significant area where Writer actually beats Word yet. In contrast, Writer lags behind Word enormously in usability, in robustness and, for some features, in power as well.
In other words, what keeps Microsoft selling Word is the fact that Word remains by far the best word processing software available for Windows. (Mac users might like to try Apple's Pages.) In time, the competition may catch up, but if users do migrate, it won't be just because someone worked out how to convert reliably between.doc and.sxw files.
You shouldn't have to choose between security and use of your computer.
True.
Because ActiveX is insecure, developers ought to use other tools to provide both compatibility with other browsers and operating systems on which ActiveX hasn't been implemented.
False, on at least two counts.
Firstly, ActiveX isn't inherently insecure: there's nothing unsafe about using known ActiveX controls from the corporate intranet web pages we create ourselves (at least, no more than running anything else over the corporate network). The permissions and security system for ActiveX is inadequate when used with untrusted content, but that's not the issue we're addressing here.
Secondly, why should the developers cater to multiple browsers and operating systems? Windows XP is our corporate standard desktop OS, and it's assumed that all corporate staff will be using it. If you want to use an alternative to the officially supported product, it's up to you to make sure that the functionality is sufficient.
You can argue that it was a poor choice of corporate standard, and perhaps you're right, but it is the standard they've chosen. Since supporting anything else costs time and money, they're hardly going to do it just because you or I happen to like Firefox.
Bottom line: if you're going to evaluate a product's pros and cons critically, as the subject of this discussion suggests, then you have to be realistic and admit that not interoperating with existing standards is a huge con.
Most people consider the lack of ActiveX a good thing, as it strengthens security considerably.
Some consider it a good thing. Others -- notably large numbers whose company intranets rely on ActiveX tools -- consider it a pain in the ass, and any security implications a distant second in importance to keeping those tools working. This is the single biggest blocker I've seen to widespread adoption of Firefox in the many workplaces.
You should fix your applications. You'll need to eventually, anyway
Why?
I am a Firefox fan, BTW; I use it both at home and work. I'm just a realist as well, and one whose office intranet uses file:// links and ActiveX scripting all over the place.
The issue is that most word processors do hard code fonts.
[Glances at Word document]
[Glances at OpenOffice.org Writer document]
Are you sure?
(Yes, many WPs can embed fonts in document files if requested, but I can't think of a single one that does it by default. And if you're using proprietary fonts that are covered by legal restrictions on distribution in your jurisdiction and you deliberately tell your WP software to embed them for distribution anyway, well, isn't that your problem?)
More security often means stuff with bad security breaks.
If stuff with bad security were all that broke after installing SP2, that would be fair enough... but it just ain't so.
For a start, taking out a previouly stable system is hardly a security fix. There have been numerous cases reported where this happened, often requiring a reinstallation of the entire OS.
Then there are numerous cases of broken hardware support, where devices that previously worked fine no longer do -- and not just networking stuff, but USB devices, hard drive subsystems, and more. How can you class any such failure as a security fix?
I find it hard to believe that all of the software that is known to break does so because either it was a security risk or it violated Microsoft's interface standards, too. It's the difference between breaking something like a poorly written personal firewall and breaking something like a completely non-Internet-enabled graphics app. SP2 has plenty of victims in both categories.
You (And well virtually all of SlashDuh) talk as if Linux has never had a single bug or security problem.
Hmm... I must have forgotten the part where I said anything about Linux (an operating system I don't often use).
MS is doing a lot to overcome the same problems that EVERY OTHER OS HAS.
No, it's doing a lot to overcome FAR MORE problems than ANY other OS has, and the fact that it has more problems to overcome is a direct consequence of not prioritising security sufficiently during earlier development. This is a self-inflicted wound. All major OSes have security issues, but most aren't practically trying to commit suicide.
And yet here, they can do nothing right. EVAR.
When they do things right, I say so. When they screw up, I say so. Any supposed security update that can break vast numbers of things totally unrelated to any security vulnerability is a screw-up.
Excuse me while I uninstall and reinstall Firefox for a third time to patch it's latest exploit...
Go ahead. Personally, I find just letting it automatically update patches everything just fine, and has done for quite a while now.
FWIW, I can second this. The local computer store we bought our office PCs from for a while happened to be installing a video card with a second port. Once when we upgraded a few people there were some spare 17" monitors lying around, and one of the guys tried hooking it up to his XP box to see how Windows coped with a second screen. The answer was that it coped very well, as did the applications in almost all cases, and he (and now several other colleagues) became an instant convert. Having a real second monitor makes using virtual desktops feel like the cheap hack it basically is.
Those of us who got stuck with corporate standard clone Dell boxen, which have a really cheap video card without the second port, are suitably impressed.;-)
A British person suing another British person for actions taken from British soil? I'd try the UK.
But the most important thing is to get the offending content removed, and that's based in the US. A British court can't order a US-based company to take down the post, and as so many people have so thoughtfully reminded us throughout this thread, ultimately it could require a court order to make them do so.
I'm sure that's been proved by someone doing something online that they could have done offline just as well.
But with all of the problems I listed, including this one, the point is that they can't be done off-line just as well. There is no other remote, wide-distribution, instant-access communications medium in existence that is available to individuals.
In any case, you've repeatedly said that you don't know the first thing about what the laws are
Once again you infer something that I haven't actually said. I've said I don't know how to pursue a legal remedy in this case, because of the complications of jurisdiction. I'm well aware of what the basic provisions in law are, at least in my own country, and of what action I would have been able to take had everyone concerned been based here.
That doesn't change the facts that (a) since we're not all in my country, that approach isn't likely to work (see my first answer above), and (b) the remedies available in current law still aren't adequate, for the reasons I described before.
Must be great to be a decision maker at Microsoft where whatever choices you take it won't be liked.
The no-win decision is a direct result of their earlier decisions, which in turn were based on boosting profits at the expense of developing good software. You'll find no sympathy here.
I recently obtained a copy of Visual Studio 2005 which I wanted to play around with. Install went fine (on XP) UNTIL I tried to install the DOCUMENTATION...which insisted that XP SP2 had to be installed!!
Would you mind posting a little more detail about this, please? We're shortly going to be in a similar position: needing to use VS2005 to match clients, but corporate standards forbid installing SP2.
In particular, which version of VS2005 are you using (beta 1, one of the CTPs?), and which documentation claimed it needed SP2?
The only VS2005 installation I've played with so far was a beta 1 version of VC++ on a 64-bit box, which seemed to go OK, but I didn't install it myself so maybe there were hurdles to leap along the way.
Slashdot Mirror
I've heard that a lot, and certainly the argument has merit.
Where it sometimes breaks down, IMHO, is that learning a new language isn't free. It's a myth that any good programmer can learn a new language in a week. Sure, they can learn the basic syntax, and if they're familiar with the particular paradigms they're using (by which I mean OOP, functional programming, whatever) then they'll be able to apply those principles fairly readily. But there's a world of difference between that and the kind of clean, idiomatic, easily maintainable code than a good programmer with a lot of experience in the specific language(s) he's using would write.
I predict that this is going to be one of the bigger factors holding back $SCRIPTING_LANGUAGE from wider usage for a long time. There are simply too many almost isomorphic scripting languages with a significant, but still small, user base, and while they offer similar advantages, you can't just switch a whole dev team from one to another for the reason above.
I think this is also a big reason C++ remains popular, particularly with stronger programmers who make the effort to learn its idioms, who for some probably related reason rarely seem to encounter all these dreadful weaknesses C++ is supposed to have... In a nutshell, C++ is a single language with sufficient tools to work effectively with both low-level details and high-level designs, and perhaps a strong programmer with such a tool will be more effective than a programmer with two more suitable tools, but only moderate skill with either.
A fairer question might be what's right about it. ;-)
Seriously, the string-handling in standard C++ suffers horribly from design-by-committee.
To give the most obvious example, there's no way to write a literal std::string, which in turn means that basic syntax like "a"+"b" doesn't work, even though the + operator is used to concatenate std::string objects.
There are also glaring inconsistencies in usage, because of all the history using char*, and then numerous class libraries inventing their own string type. Take a look at the I/O streams interfaces in C++: you can't even pass a std::string as a filename without converting to a C-style string first.
Then there's the interface, which is a bit Jack-of-all-trades, and bloated with it.
And of course, even with a real string type now -- surely a prerequisite for claiming any meaningful string support at all! -- C++ still doesn't provide standard support for basic string processing using regular expressions. Compared to a more text-focussed language like Perl, well, there's no comparison.
Fortunately, a number of these sillies are in line to be fixed in the next revision of C++, but for now, the language's string support is pretty much a joke.
Just out of interest, how many such "glaring warnings" do you think a national commander-in-chief receives during his career? How many of them do you think appear so glaring before they become reality?
Do you really think POTUS can't give any orders he feels necessary immediately, even when he's on vacation?
If you think he misjudged the briefing, fine. If you think he did nothing when he should have acted, fine. But "coming back from vacation" is just PR.
In many cases, yes it is. In some cases, it's not, and the work of aid agencies that rely primarily on charitable contributions to fund their work is critical. Just look at what's been achieved near the tsunami site, where the public collectively acknowledged the problem and voted with their wallets to do something about it. Yet more people die in Africa every year than died as a result of the tsunami, and we do relatively little to help.
We (several first world countries) have just overthrown the governments of both Afghanistan and Iraq for far less constructive purposes. Do you really think anyone at the top of, say, the US or UK governments is ever going to be held accountable? Hell, the US re-elected Bush despite that, and sadly it looks as though we in the UK are about to re-elect Blair (though probably with a much-reduced majority, not least because of his actions relating to Iraq).
Some of the stuff that goes on in Africa makes the Middle East dictatorships look pleasant, and there's an obvious humanitarian motivation for waging war. Of course, they don't produce as much oil, and presumably it would therefore be somehow more illegal under international law to topple people like Mugabe than it was to topple Hussein or the Taleban?
And how many memos do you think the most powerful man in the world gets every day, trying to draw his attention to some concern that the author thinks is the most important thing in the world?
Some groups had been advocating a tsunami warning system for years, which would have saved orders of magnitude more lives than foreseeing 9/11, at a fraction of the cost of clearing up the damage the tsunami caused when it hit.
There are enough resources in the world to feed every human alive a good diet, yet many thousands die through malnutrition every year in third world countries, because the rest of the world doesn't do enough to help them.
If a polar ice cap melts next year, causing millions of deaths as coastal cities are flooded all over the world, the critics will be saying that the environmentalists had been warning of this for years.
If the super-volcanic eruption happens and blacks out the whole US, also killing millions, someone will point out that there was a documentary on TV a few months ago warning about the possibility.
If an unpredicted asteroid hits the earth tomorrow and wipes out half the population of the planet, someone will be saying that the warnings were there and billions should have been invested in a super-defence-network-satellite-system-thing.
I don't know whether Bush really missed something he should have seen, or indeed if the intelligence community collectively did. Maybe so, in which case that's a problem that needs fixing. But, while I'm no fan of GWB, I do have realistic expectations, and I recognise that there are limits to what one man -- however powerful -- can see and do.
Are you sure it wasn't just a made-up concept with a pseudo-funky name that the new designer invented to sound "cool"?
Half of that redesign wasn't a change in graphic design, it was a complete change in content and implied a complete change in process. Chances are the designer had absolutely no idea what processes this document actually fits into in reality, and has equally little idea whether the changes in content he made would offer any value, or even whether they'd actually make the document less useful.
We should have just gone to red alert, switched on the force fields, and tractor-beamed the planes into the docking bay, Captain.
Oh dear. As I read this you're at +5. Guess you're screwed, then. ;-)
Probably as far as the early hours of 6 May. After that, all bets are off. ;-)
Whatever you think of his politics, he's a very shrewd operator. He's long been a close friend of [UK Prime Minister] Tony Blair, and despite the bad press (justified or otherwise) that some of his actions have drawn, he's continued a high-flying political career. His resignations have avoided a lot of the amateurish fall-out that some of his peers (David Blunkett and Clare Short, for two examples) incurred by messing up the PR, by taking him out of the limelight just long enough for the public to forget any transgressions he might have committed. Of course, they also get him nicely spoofed on the political satire shows, but such is the price of celebrity. ;-)
I understand your frustration, but it sounds like the original code wasn't written in standard C++ when it could have been; the standard has been around since about '97 for practical purposes, after all. This is a great example of why we have standards: if you write code to the standard, it will compile on any standard-compliant compiler (yeah, I know), and it will continue to compile on any future standard-compliant versions. If you rely on non-standard features or implementation-defined or undefined behaviour, then you're trusting to fate whether your code will continue to compile and work the same way in future.
In other words, this is certainly a real problem, but I think you're attributing the blame to the wrong source. When you write
you imply that the older code wasn't broken already, which probably isn't the case -- certainly if it was written as recently as 2002!
FWIW, I work with a lot of code that was written pre-standard, and the pre-standard compilers to match, every day, and it can be very frustrating when things that are trivial in standard C++ take extra code clutter and testing time to achieve "just because". I feel your pain! :-)
So I've been told.
Unfortunately, in the real world, porting that technology isn't free: at the very least, it takes time and resources. And since those extensions aren't standard, it may well not be the best use of whatever limited time and resources other compiler writers have.
Moreover, the fact that GCC is GPL'd makes lifting its algorithms straight out a massive no-no for any commercial compiler vendor I can think of. Without wanting to restart the usual pointless holy war, this is an obvious case where the GPL being free-as-in-FSF and not free-as-in-completely may well be inhibiting reuse of existing code/design work.
This all sounds a bit harsh, so if you can name me any other C++ compiler that effectively implements all the C++ extensions used in the major Linux-related projects, I'll concede that the reuse isn't being completely inhibited.
Damn. You have to compile it? No wonder mine hasn't loaded yet...
That's fine, but what if the amount of time required to list those requirements accurately could be 80% of the total development time for the project? Anyone can give you an estimate for the release date five minutes before they release the code. ;-)
Seriously, I see this all the time at the moment. I'm adding new features to a relatively old piece of software. My last project was timetabled to take about six weeks of development, but lasted nearly a year. The new development work I did actually took (give or take a couple of days) the amount of time I'd estimated, and met its spec after that time. The problem was that then they found a zillion bits of old code that didn't meet their spec, hadn't been tested sufficiently to discover this a decade ago, but now weren't up to the job of driving my nice new functionality. Fixing those extended the length of the project as a whole by an order of magnitude.
Now, I could have told the project managers that was going to happen. But until I'd spent the several months I took investigating the older code to identify the numerous special cases that would need removing in order to reduce it to a systematic, effectively testable design, I couldn't have told them how long it was going to take. Once you've got that knowledge, writing the code itself is barely significant in the amount of time it takes.
Sure. Random, untrained users often fail to take advantage of even the basic features of their software. As the saying goes, if you think training is expensive, try ignorance.
However, those who produce documents for a living and take an hour or two to learn the basics of word processing at some point in their lives will gain better results faster by using stylesheets, automatic mail merge, automatic cross-references and the like. Actually, those are three of Word's weaker features, but even then it remains some way ahead of OpenOfficer Writer, at least for now.
I was going to mod you down for trolling, but I'd rather set the record straight.
There is clearly more to new versions of Word than being able to open previous versions' documents. If there weren't, then far more businesses would have given up upgrading by now. The average home user may not use all (or perhaps even any) of the new features, particularly in the most recent incarnations where they've been heavily focussed on interoperability with other Office apps, but many business users do.
Even without the new features, Word remains a far more powerful tool than OpenOffice.org Writer (and I say this as someone who works with both apps regularly, on a wide range of document types). OO.o has enormous potential, and I'm grateful to the dev team for giving me such a useful product that I can install for free at home, but let's keep some perspective. I can't think of any significant area where Writer actually beats Word yet. In contrast, Writer lags behind Word enormously in usability, in robustness and, for some features, in power as well.
In other words, what keeps Microsoft selling Word is the fact that Word remains by far the best word processing software available for Windows. (Mac users might like to try Apple's Pages.) In time, the competition may catch up, but if users do migrate, it won't be just because someone worked out how to convert reliably between .doc and .sxw files.
True.
False, on at least two counts.
Firstly, ActiveX isn't inherently insecure: there's nothing unsafe about using known ActiveX controls from the corporate intranet web pages we create ourselves (at least, no more than running anything else over the corporate network). The permissions and security system for ActiveX is inadequate when used with untrusted content, but that's not the issue we're addressing here.
Secondly, why should the developers cater to multiple browsers and operating systems? Windows XP is our corporate standard desktop OS, and it's assumed that all corporate staff will be using it. If you want to use an alternative to the officially supported product, it's up to you to make sure that the functionality is sufficient.
You can argue that it was a poor choice of corporate standard, and perhaps you're right, but it is the standard they've chosen. Since supporting anything else costs time and money, they're hardly going to do it just because you or I happen to like Firefox.
Bottom line: if you're going to evaluate a product's pros and cons critically, as the subject of this discussion suggests, then you have to be realistic and admit that not interoperating with existing standards is a huge con.
Some consider it a good thing. Others -- notably large numbers whose company intranets rely on ActiveX tools -- consider it a pain in the ass, and any security implications a distant second in importance to keeping those tools working. This is the single biggest blocker I've seen to widespread adoption of Firefox in the many workplaces.
Why?
I am a Firefox fan, BTW; I use it both at home and work. I'm just a realist as well, and one whose office intranet uses file:// links and ActiveX scripting all over the place.
[Glances at Word document]
[Glances at OpenOffice.org Writer document]
Are you sure?
(Yes, many WPs can embed fonts in document files if requested, but I can't think of a single one that does it by default. And if you're using proprietary fonts that are covered by legal restrictions on distribution in your jurisdiction and you deliberately tell your WP software to embed them for distribution anyway, well, isn't that your problem?)
If stuff with bad security were all that broke after installing SP2, that would be fair enough... but it just ain't so.
For a start, taking out a previouly stable system is hardly a security fix. There have been numerous cases reported where this happened, often requiring a reinstallation of the entire OS.
Then there are numerous cases of broken hardware support, where devices that previously worked fine no longer do -- and not just networking stuff, but USB devices, hard drive subsystems, and more. How can you class any such failure as a security fix?
I find it hard to believe that all of the software that is known to break does so because either it was a security risk or it violated Microsoft's interface standards, too. It's the difference between breaking something like a poorly written personal firewall and breaking something like a completely non-Internet-enabled graphics app. SP2 has plenty of victims in both categories.
Blockquoth the AC:
Hmm... I must have forgotten the part where I said anything about Linux (an operating system I don't often use).
No, it's doing a lot to overcome FAR MORE problems than ANY other OS has, and the fact that it has more problems to overcome is a direct consequence of not prioritising security sufficiently during earlier development. This is a self-inflicted wound. All major OSes have security issues, but most aren't practically trying to commit suicide.
When they do things right, I say so. When they screw up, I say so. Any supposed security update that can break vast numbers of things totally unrelated to any security vulnerability is a screw-up.
Go ahead. Personally, I find just letting it automatically update patches everything just fine, and has done for quite a while now.
FWIW, I can second this. The local computer store we bought our office PCs from for a while happened to be installing a video card with a second port. Once when we upgraded a few people there were some spare 17" monitors lying around, and one of the guys tried hooking it up to his XP box to see how Windows coped with a second screen. The answer was that it coped very well, as did the applications in almost all cases, and he (and now several other colleagues) became an instant convert. Having a real second monitor makes using virtual desktops feel like the cheap hack it basically is.
Those of us who got stuck with corporate standard clone Dell boxen, which have a really cheap video card without the second port, are suitably impressed. ;-)
But the most important thing is to get the offending content removed, and that's based in the US. A British court can't order a US-based company to take down the post, and as so many people have so thoughtfully reminded us throughout this thread, ultimately it could require a court order to make them do so.
But with all of the problems I listed, including this one, the point is that they can't be done off-line just as well. There is no other remote, wide-distribution, instant-access communications medium in existence that is available to individuals.
Once again you infer something that I haven't actually said. I've said I don't know how to pursue a legal remedy in this case, because of the complications of jurisdiction. I'm well aware of what the basic provisions in law are, at least in my own country, and of what action I would have been able to take had everyone concerned been based here.
That doesn't change the facts that (a) since we're not all in my country, that approach isn't likely to work (see my first answer above), and (b) the remedies available in current law still aren't adequate, for the reasons I described before.
If it works, use it.
Or don't install SP2.
Or don't install SP2.
This one is Microsoft vs. the corporate world, and looking at the acceptance stats in TFA, the current odds are around 3:1 against Microsoft.
No-one providing a business-related service is going to require something that around 3/4 of their customers don't have; it would be business suicide.
The no-win decision is a direct result of their earlier decisions, which in turn were based on boosting profits at the expense of developing good software. You'll find no sympathy here.
Would you mind posting a little more detail about this, please? We're shortly going to be in a similar position: needing to use VS2005 to match clients, but corporate standards forbid installing SP2.
In particular, which version of VS2005 are you using (beta 1, one of the CTPs?), and which documentation claimed it needed SP2?
The only VS2005 installation I've played with so far was a beta 1 version of VC++ on a 64-bit box, which seemed to go OK, but I didn't install it myself so maybe there were hurdles to leap along the way.