We are pragmatic in accomplishing our goals and we don't want to muck about with the tools.
But using BitKeeper wasn't pragmatic, though, and that's the whole point. It's not like tens of thousands of the people who follow such events weren't screaming that this was a Bad Idea, and their worst fears came to pass.
Then again, when faced with the task of creating a web frontend to a FoxPro database, my first project was to develop a SOAP interface to Windows ADO databases so that I could write the rest of the project on FreeBSD instead of Windows. The short term "pragmatic" concept would've been to base the whole project on Windows, but that wasn't an acceptable long-term solution. I don't really understand the short-term-oriented decision process (nor do I particularly want to).
Never occured to me that a company might leverage a market position with a crappy metadata format.
Goodbye, sweet innocence... Yeah, I've been there. I just don't want to make that mistake again, especially by rationalizing it to myself as a practical solution.
By the way, I'm very pro-capitalist, too. I just think that we're dealing with a very artificial market right now, and that commodities like operating systems and word processors will inevitably become free (and hopefully Free). There'll always be room for proprietary in-house development, and that's why I have no fear of the upcoming change. I'm fortunate enough to have a job already that pays me to write Free software to internal use but I expect that will become the norm.
Nothing is black and white, there's plenty of shades of gray and that's exactly where everybody, but extremists, are.
Sorry, but you're wrong. Making the conscious decision to give up your freedom to avoid a little inconvenience isn't a shade of gray. Unless you have a very specific interoperability need, choosing proprietary software to do a job that could be accomplished by Free code isn't good, logical, or pragmatic.
I can understand people who make that choice out of ignorance, since if you don't know something exists then you can't be expected to consider it. Linux certainly isn't one of those people, though, and RMS was right to say so.
I wasn't always an RMS fan. To tell the truth, he seemed like a ranting nutjob when I first switched to GNU/Linux way back when. However, every single time I've dismissed his radical doomsayings, later events have shown that he was deadly accurate. Every time. Always. You can only have so many "that could never happen - what? It did?" moments before you understand why the man is so frustrated. It'd be a cruel life indeed to be so consistently and publicly vindicated but still have people write you off.
Thus I'd have two options: make the software free from the start (donating the programming effort with no gain) or not write it at all.
Tell you what, when someone comes out with program for GNU/Linux that's developed entirely without Free tools (read: no GPL or BSDed editors, compilers, libc, make, and so on), then I'll be a little more sympathetic to that position. Until then, the concerns of people who want to build proprietary apps using the work of Free developers - even if they're legally entitled to do so - are absolutely unpersuasive to me.
That's not impossible, either. Someone could quite conceivably write a Linux app entirely via another OS's IDE, statically compile it with the Intel or Microsoft compilers, and only move it to a Linux box when it's ready to test. Short of that, though, it comes down to someone wanted to use the huge body of work written by GNU and other Free contributors without giving back to the community. Why should I even care what people like that think?
Sometimes it makes more sense to buy something because it is supported and stable and someone can be held accountable for mistakes.
Those rationales make me want to slap someone. I've never used proprietary software where:
Tech support was anywhere near the quality of the Free equivalents.
The vendor was willing to be held accountable.
As much as Didio types love to shrill about support costs, the FreeBSD, Debian, and Gentoo mailing lists (and newsgroups and web forums and IRC channels) are much faster and more responsive than any commercial avenues I've ever had the displeasure of using.
And as far as the whole "accountability" idea, good luck getting Microsoft to accept the blame for the latest exploited vulnerability. Even if you do manage to deflect blame to them in front of your management, at some point your bosses are going to want to know why you keep choosing that vendor when their track record is so terrible.
I don't really mean to rant at you - I know you're just echoing some of the most common reasons for choosing proprietary systems - but those excuses still make my skin crawl.
Boy, if that isn't the truth. Considering how many people are quick to paint RMS as an extremist, I think it's ironic that he's far more pragmatic than 99.99% of them. Willingly locking yourself into someone else's game when there are other alternatives around (even if somewhat less technically featureful) is not a reasonable or even practical thing to do, but the Open Source advocates seem willing to experience the lesson time and again without actually learning the principle.
Controller for an MRI scanner? Proprietary is OK. Microcode for an anti-lock braking system? Proprietary is OK. Your company's business logic, web services, email, word processing, version control? Free alternatives exist - proprietary is not OK. That's the pragmatic answer, which just happens to correspond with the ideological one.
If you don't mind living at the whims of a third party who rarely has your best interests in mind, then maybe gratis isn't such an irrational choice. If you want to own your own data, though, then libre trumps gratis in every single case I've ever come across.
What's the point of running a high resolution if you only have one window open?
Setting large fonts and placing the monitor far enough away that I have nice, sharp text at a comfortable viewing distance. It works perfectly except when certain web designers want to restrict parts of their layout to an absolute rather than relative size.
Between the Click-o'-death, Jaz WORN drives (Write Once Read Never), a handful of short-lived proprietary formats, and an unhealthy denial that any of the above ever happened, you couldn't pay me to store my important data on anything made by Iomega. They're like the Yugo of storage devices, and the Chicago Cubs of backup media (except that people like the Cubs).
On the other hand, their products work great as entropy sources for hardware PRNG acceleration.
For example text areas on my contact form that have a max-width of six inches stretch the entire way across the page in Internet Explorer
I'm glad to see Internet Explorer doing something right, even if inadvertently. See, I have this nice 19" monitor, and people who insist on making tiny little pages that fill the top-left corner of my screen make me leave their site as quickly as possible. I spent good money to have a lot of screen real estate - please don't try to take it away from me.
I've already upgraded my friends and relatives. My office has completely switched (except for a person or two that relies on years' worth of VB automation). Our local city government is considering it. My oldest kid's school is looking into it and talking about handing out CDs to parents.
People are flocking in droves to OpenOffice. It just doesn't get as much press as Firefox.
Unfortunately, there is a BIG learning curve to teach the unsavvy anything other than what they are used to.
The article didn't mention which version of Office he'd be upgrading from. If it's something moderately old (maybe Office 97, which would be pretty reasonable given the governmental nature of the job), then I'd say that Office XP will require as much training as OpenOffice.
On the same note, my 45-person company was facing a group upgrade from Office 97, and our enlightened IT guy switched everyone to OpenOffice at that point. After the first week everyone just took it for granted and never really mentioned it again.
if there are several machines on a network, and I set up printing by attaching my USB laser printer to one of them and running the KDE printer-setup program, the other machines then see it automatically, no work necessary on my side.
I love that feature. It makes it trivially easy to print on the Epson Photo Stylus on my wife's iMac from my FreeBSD machine, and for her to print to the Laserjet on mine. In fact, that's usually how I find out that I have printer problems: she can't print to my machine until I go fix it.
I don't care how fast a processor is--I usually have one task that will crush it--but rarely do I have two time-critical things to worry about at the same time.
Maybe that's because you've been trained not to run two intensive applications at the same time. If your subconscious is telling you not to launch a compile job until your DVD transcode is done, then you probably wouldn't get a lot out of an SMP system. Break that habit, though, and you might like what you find.
The Doppler radar that your local TV station bought and raves about is completely useless for forecasting
I agree with everything but that. I take it you don't live in Tornado Alley, where even 5 minutes of notice that a storm has developed a hook is enough to save quite a few lives. Doppler won't help a bit with 7-day forecasts, but it's really really nice to be able to see exactly where the bad parts of an approaching storm are relative to where I am so I can forecast whether I'm going to die within the next 10 minutes.
Now, I use -Os instead, which is like -O2 except it doesn't add certain binary-bloating flags that are present in -O2.
From the same info page:
`-Os' disables the following optimization flags:
-falign-functions -falign-jumps -falign-loops
-falign-labels -freorder-blocks -fprefetch-loop-arrays
IANA compiler expert, but it looks like those would be pretty good optimizations to leave enabled, even if the increase the object size somewhat - as long as you're not complete starved for cache. What CPU do you use -Os on? Have you benchmarked the differences?
I have an older Alpha that I might give -Os a test on whenever I can afford to let it grind away at a compile for a few hours.
Optimize yet more. `-O3' turns on all optimizations specified by
`-O2' and also turns on the `-finline-functions', `-fweb' and
`-frename-registers' options.
Is anyone using "-O2 -fweb"? It seems like the sort of 1337 thing that someone would redundantly toss in with -O99 or the like.
But easier then windows? I bought a printer, plugged it in and it worked. Never took the driver out of the box.
Yes, I'm serious. It's no harder than Windows in the most common situation where the printer in question is supported and attached locally, but a whole lot easier than Windows when the printer is somewhere else on the network. With autodiscovery, there's a pretty good chance you don't even have to do anything beyond selecting which printer you want to use among the ones currently available when the print dialog was opened. It doesn't get much easier than that.
look, we'd really love to keep you on but I'm afraid there's no legal protection to you if we fire you over your homosexuality, so we're forced to let you go.
In other words, "we need to preemptively fire you for your own protection"? I call BS. Either you're leaving out something, or you were on your way to being fired for a different reason and he was giving you a legal excuse.
I can't believe MS is really afraid of a religious-right boycott, especially when they're still the darlings of the other side of the Republican party (the economic right).
You're really generalizing there. I'm literally a card-carrying Republican, largely for economic reasons, but I certainly have no love for the convicted illegal monopolists. Liberals are not a homogenous block, and neither are conservatives. Just thought you'd like to know.
I could see that as a substitute for civil unions if it were available to both orientations and on-demand. Otherwise, a straight couple could make a pretty reasonable argument that they shouldn't have to wait a year (or however long it takes to be married under common law in Washington) when a gay couple can sign a piece of paper giving them the same privileges immediately.
Which leads to a related question: if John and Kevin apply for things normally only available to married people by signing an affidavit that they're partners, do they have to go through divorce proceedings to dissolve the arrangement? Do John and Jane married under common law? That's not a leading question; I honestly don't know the answer.
As I understand it, it has nothing to do with relationships, just with sexual orientation.
I don't think that's true. Regarding insurance: a gay man could marry a gay woman (see Andrea Dworkin for a real-life example of that) and receive the exact same policy as a straight man married to a straight woman. The new proposed law would extend those same benefits to gay couples; single gay people would not benefit from that part of the law.
OK. Is there a form for straight unmarried partners to sign so that they can receive the same benefits?
The schism really seems to be between married and unmarried couples, rather than gay and straight couples. I think governments would be better served to address that issue than to patch a bunch of smaller laws.
Slashdot Mirror
But using BitKeeper wasn't pragmatic, though, and that's the whole point. It's not like tens of thousands of the people who follow such events weren't screaming that this was a Bad Idea, and their worst fears came to pass.
Then again, when faced with the task of creating a web frontend to a FoxPro database, my first project was to develop a SOAP interface to Windows ADO databases so that I could write the rest of the project on FreeBSD instead of Windows. The short term "pragmatic" concept would've been to base the whole project on Windows, but that wasn't an acceptable long-term solution. I don't really understand the short-term-oriented decision process (nor do I particularly want to).
Never occured to me that a company might leverage a market position with a crappy metadata format.
Goodbye, sweet innocence... Yeah, I've been there. I just don't want to make that mistake again, especially by rationalizing it to myself as a practical solution.
By the way, I'm very pro-capitalist, too. I just think that we're dealing with a very artificial market right now, and that commodities like operating systems and word processors will inevitably become free (and hopefully Free). There'll always be room for proprietary in-house development, and that's why I have no fear of the upcoming change. I'm fortunate enough to have a job already that pays me to write Free software to internal use but I expect that will become the norm.
Sorry, but you're wrong. Making the conscious decision to give up your freedom to avoid a little inconvenience isn't a shade of gray. Unless you have a very specific interoperability need, choosing proprietary software to do a job that could be accomplished by Free code isn't good, logical, or pragmatic.
I can understand people who make that choice out of ignorance, since if you don't know something exists then you can't be expected to consider it. Linux certainly isn't one of those people, though, and RMS was right to say so.
I wasn't always an RMS fan. To tell the truth, he seemed like a ranting nutjob when I first switched to GNU/Linux way back when. However, every single time I've dismissed his radical doomsayings, later events have shown that he was deadly accurate. Every time. Always. You can only have so many "that could never happen - what? It did?" moments before you understand why the man is so frustrated. It'd be a cruel life indeed to be so consistently and publicly vindicated but still have people write you off.
Tell you what, when someone comes out with program for GNU/Linux that's developed entirely without Free tools (read: no GPL or BSDed editors, compilers, libc, make, and so on), then I'll be a little more sympathetic to that position. Until then, the concerns of people who want to build proprietary apps using the work of Free developers - even if they're legally entitled to do so - are absolutely unpersuasive to me.
That's not impossible, either. Someone could quite conceivably write a Linux app entirely via another OS's IDE, statically compile it with the Intel or Microsoft compilers, and only move it to a Linux box when it's ready to test. Short of that, though, it comes down to someone wanted to use the huge body of work written by GNU and other Free contributors without giving back to the community. Why should I even care what people like that think?
Those rationales make me want to slap someone. I've never used proprietary software where:
As much as Didio types love to shrill about support costs, the FreeBSD, Debian, and Gentoo mailing lists (and newsgroups and web forums and IRC channels) are much faster and more responsive than any commercial avenues I've ever had the displeasure of using.
And as far as the whole "accountability" idea, good luck getting Microsoft to accept the blame for the latest exploited vulnerability. Even if you do manage to deflect blame to them in front of your management, at some point your bosses are going to want to know why you keep choosing that vendor when their track record is so terrible.
I don't really mean to rant at you - I know you're just echoing some of the most common reasons for choosing proprietary systems - but those excuses still make my skin crawl.
Controller for an MRI scanner? Proprietary is OK. Microcode for an anti-lock braking system? Proprietary is OK. Your company's business logic, web services, email, word processing, version control? Free alternatives exist - proprietary is not OK. That's the pragmatic answer, which just happens to correspond with the ideological one.
If you don't mind living at the whims of a third party who rarely has your best interests in mind, then maybe gratis isn't such an irrational choice. If you want to own your own data, though, then libre trumps gratis in every single case I've ever come across.
The irony is not lost on this previous Commodore fan: "Amiga? Those are good for games, maybe, but you need DOS to get real work done."
The correct answer, as always, is "my new religion forbids it". There's just not much AOL/Cox/Soldier OF Fortune can come back with to counter it.
Setting large fonts and placing the monitor far enough away that I have nice, sharp text at a comfortable viewing distance. It works perfectly except when certain web designers want to restrict parts of their layout to an absolute rather than relative size.
On the other hand, their products work great as entropy sources for hardware PRNG acceleration.
I'm glad to see Internet Explorer doing something right, even if inadvertently. See, I have this nice 19" monitor, and people who insist on making tiny little pages that fill the top-left corner of my screen make me leave their site as quickly as possible. I spent good money to have a lot of screen real estate - please don't try to take it away from me.
People are flocking in droves to OpenOffice. It just doesn't get as much press as Firefox.
The article didn't mention which version of Office he'd be upgrading from. If it's something moderately old (maybe Office 97, which would be pretty reasonable given the governmental nature of the job), then I'd say that Office XP will require as much training as OpenOffice.
On the same note, my 45-person company was facing a group upgrade from Office 97, and our enlightened IT guy switched everyone to OpenOffice at that point. After the first week everyone just took it for granted and never really mentioned it again.
I love that feature. It makes it trivially easy to print on the Epson Photo Stylus on my wife's iMac from my FreeBSD machine, and for her to print to the Laserjet on mine. In fact, that's usually how I find out that I have printer problems: she can't print to my machine until I go fix it.
Maybe that's because you've been trained not to run two intensive applications at the same time. If your subconscious is telling you not to launch a compile job until your DVD transcode is done, then you probably wouldn't get a lot out of an SMP system. Break that habit, though, and you might like what you find.
I agree with everything but that. I take it you don't live in Tornado Alley, where even 5 minutes of notice that a storm has developed a hook is enough to save quite a few lives. Doppler won't help a bit with 7-day forecasts, but it's really really nice to be able to see exactly where the bad parts of an approaching storm are relative to where I am so I can forecast whether I'm going to die within the next 10 minutes.
From the same info page:
IANA compiler expert, but it looks like those would be pretty good optimizations to leave enabled, even if the increase the object size somewhat - as long as you're not complete starved for cache. What CPU do you use -Os on? Have you benchmarked the differences?
I have an older Alpha that I might give -Os a test on whenever I can afford to let it grind away at a compile for a few hours.
That makes a little more sense. Thanks for the info.
Yes, I'm serious. It's no harder than Windows in the most common situation where the printer in question is supported and attached locally, but a whole lot easier than Windows when the printer is somewhere else on the network. With autodiscovery, there's a pretty good chance you don't even have to do anything beyond selecting which printer you want to use among the ones currently available when the print dialog was opened. It doesn't get much easier than that.
Because there's no state-sponsored equivalent "certification". That sounds like the real problem people are trying to fight.
In other words, "we need to preemptively fire you for your own protection"? I call BS. Either you're leaving out something, or you were on your way to being fired for a different reason and he was giving you a legal excuse.
You're really generalizing there. I'm literally a card-carrying Republican, largely for economic reasons, but I certainly have no love for the convicted illegal monopolists. Liberals are not a homogenous block, and neither are conservatives. Just thought you'd like to know.
Which leads to a related question: if John and Kevin apply for things normally only available to married people by signing an affidavit that they're partners, do they have to go through divorce proceedings to dissolve the arrangement? Do John and Jane married under common law? That's not a leading question; I honestly don't know the answer.
I don't think that's true. Regarding insurance: a gay man could marry a gay woman (see Andrea Dworkin for a real-life example of that) and receive the exact same policy as a straight man married to a straight woman. The new proposed law would extend those same benefits to gay couples; single gay people would not benefit from that part of the law.
The schism really seems to be between married and unmarried couples, rather than gay and straight couples. I think governments would be better served to address that issue than to patch a bunch of smaller laws.