Try SQLAlchemy. It's simply awesome. The ORM by and for people who actually know relational databases. No "composite keys not supported". No "Legacy DB" FUD phrasing like Rails. Use it with Mako for templates and whatever web framework. Pylons/Pyramid works, as do others I'm sure, but to me the far more important choice is the ORM and templating system. The rest is just glue.
Ubuntu has had a strange mix of decisions lately. Good decisions to dump f-spot and evolution. These were two default programs I always had to remove for non-technical users that I support to avoid endless complaints. Then there are bad decisions like pulseaudio (over jack). It causes problems but, after a lot of growing pains, mostly works for the average user. Then there are the horrible decisions like Unity and Wayland that make you wonder where they think they are going.
> Some of us have been writing C in Linux for ~12 years and decided enough is enough. C++ is not any better (in many ways worse),...
And that's where you went wrong. Good, modern C++ is much better than C for most high level applications. Most of the anti C++ attacks from below (C) and above (Java/C#) are merely FUD in my experience. Programming in C++ provides nice abstractions that C lacks, without giving up system-level control that the JVM/CLR environments do.
Now when will we see the Firefox Javascript JIT compiler ported to AMD64? It's not very well documented but 64-bit Firefox builds simply ignore the javascript.options.jit.content setting. It took me a while to figure out why the much promoted feature had no affect on any Javascript benchmarks.
Unfortunately tracemonky still doesn't work on 64-bit builds. You can set the javascript.options.jit.content preference to true with no warning that you are at most going to get a placebo effect on javascript performance. So anyone with a modern system won't be able to take advantage of one of the biggest new features.
Days later, the download page still links to 2.4.0 for English PPC Mac. You can get OO 3.0 in Japanese and French though.
I tried it on a Intel Mac and it works pretty well. Didn't find any bugs for a couple days. The first bug was fairly serious though. It gave an bogus error when I tried to save my file. Even Save As was broken. I was finally able to Save As a Word.doc file, re-open, and re-save as the original ODT. Later, after reducing the font size I got a few glitches in the font display, required toggling zoom levels to fix.
But overall much faster than NeoOffice, and much better than earlier attempts on Mac OS X.
A little knowledge is a dangerous thing
on
Fire Your IT Boss
·
· Score: -1
On the other hand, if your boss is able to just barely write something with PHP and MySQL, he might make you actually use these worst-in-class choices!
No one is going to to offer HD-DVD compatibility. It is dead, why would any CE company want to reduce profits by including that support? Slashdot's "collective wisdom" about this whole format war has had an abysmal track record. From people claiming there wasn't a significant difference in quality on the new formats, to the bogus claims that Warner and Toshiba were paid off by Sony, and now the bitter HD-DVD fans and Microsoft shills claiming Blu-ray is just as dead and that downloads are the new thing. The silly thing is that with even a small amount of research the outcome of the war was a foregone conclusion.
I like reiserfs because I can trust it to perform well on any file system load. I can put it on a server and know it will be fast and efficient regardless of what the users do. Ext3 gives ext2 journaling, but does not add efficient large directories or small files, two features that reiserfs has.
Sure ext3 may benchmark slightly faster in certain scenarios. But unless you know ahead of time that those are the only scenarios you are going to put on the file system, I recommend reiserfs.
I simply cannot fathom what's in your confused mind. You lambast the people who choose to take advantage of the very freedom you praise about the BSD license, all in the same breath! This is a truly absurd statement. By your logic, we should congratulate a murderer for exercising his or her freedom (ability) to purchase a weapon and end someone else's life. Or we should applaud someone who signs a note to waive their right to vote. Of course we don't want someone to sell away their freedom, however that IS their choice. One hopes that people will take advantage of their freedoms and act in that same Spirit of Freedom, however that is not always the case. The Spirit of Freedom intrinsically allows people to relinquich their freedoms. The community as a whole probably wishes everyone would work in the Spirit of Freedom. The GPL forces everyone to work like this, which is a inherent contradiction. The BSDL realizes that in the true Spirit of Freedom, you can't force someone to act in the Spirit of Freedom.
I'm not sure if I understand you, but you seem to be saying that people who BSDL their code really don't want other people to take advantage of it with a less free license. But this contradicts several of the assertions made by BSDL advocates in this forum that specifically say they don't mind if this happens.
For the record I have absolutely nothing against the BSD license. What irritates me is people who advocate the BSDL as being "more free" on the one hand, and then turn around and go ape-shit when someone wants to include BSD code in a GPL program.
I've actually seen one person suggest that a clause be added to the BSD license that specificly prohibits use in GPL code! Hey pal, you've just reinvented the GPL, moron.
> So, you're not interested in improving someone else's code, not interested in helping the community, not interested in any sort of the > customary altruistic activities inherent in a gift-culture. Instead, you just want to take.
> The author of a BSD source code can rerelease it at any time under any license he wishes. But because he doesn't, you would take > license on it against his wishes (not that you legally could). Ungrateful greedy lout!
I simply cannot fathom what's in your confused mind. You lambast the people who choose to take advantage of the very freedom you praise about the BSD license, all in the same breath!
If you were rational you would be _happy_ if someone takes BSD code and relicense it under the GPL. You would say "ah look, this exactly why I support the BSD license. Good work".
(PS. Sorry for the flame tone -- I'm in a bad mood for unrelated reasons)
> If any of the Linux people think they can just >"tack the GPL on" to the file, they are flat >wrong, plain and simple.
No, you are flat wrong, plain and simple. You are also a hypocrit. Your whole argument that FreeBSD is "more free" is that, in addition to the freedoms of the GPL, you also give the freedom to release less free versions. If Microsoft can take BSD code and include it in their less-free licensed software then the GPL people can certainly do the same.
You may be confusing this with actually changing the copyright holder on your code, in which case you are correct, but that is irrelevent to this discussion.
It was some sort of soccer game. Looked fine, but unfortunately when I asked them they admitted that they are far from having an OpenGL driver _really_ optimized for the (awesome!) fx6+ cards on Linux yet. (IE they haven't even started on DRI support yet.)
Best bet in the _near_ term for Linux is still a 3dfx Voodoo3 card, but hopefully in 6 months we'll be seeing more highend stuff (Wildcat 4000, fx6+, SGIs vpcs, Evans & Sutherland's new thing...) supported.
Also, it's not "much more difficult to port applications" because most of them are already using Unix and X, so the port is in fact easy. It's getting the "product" together -- packaging, marketing, support that takes extra effort -- for any new platform.
Again, though, not everyone can hear your internet radio station. Many many people cannot afford even what we think of as "cheap" computers and internet access fees. They can afford a cheap radio.
Slashdot Mirror
At least we will always be able to switch to T-Mobile to get competitive prices... Oh wait.
Try SQLAlchemy. It's simply awesome. The ORM by and for people who actually know relational databases. No "composite keys not supported". No "Legacy DB" FUD phrasing like Rails. Use it with Mako for templates and whatever web framework. Pylons/Pyramid works, as do others I'm sure, but to me the far more important choice is the ORM and templating system. The rest is just glue.
Ubuntu has had a strange mix of decisions lately. Good decisions to dump f-spot and evolution. These were two default programs I always had to remove for non-technical users that I support to avoid endless complaints. Then there are bad decisions like pulseaudio (over jack). It causes problems but, after a lot of growing pains, mostly works for the average user. Then there are the horrible decisions like Unity and Wayland that make you wonder where they think they are going.
> Some of us have been writing C in Linux for ~12 years and decided enough is enough. C++ is not any better (in many ways worse), ...
And that's where you went wrong. Good, modern C++ is much better than C for most high level applications. Most of the anti C++ attacks from below (C) and above (Java/C#) are merely FUD in my experience. Programming in C++ provides nice abstractions that C lacks, without giving up system-level control that the JVM/CLR environments do.
Now when will we see the Firefox Javascript JIT compiler ported to AMD64? It's not very well documented but 64-bit Firefox builds simply ignore the javascript.options.jit.content setting. It took me a while to figure out why the much promoted feature had no affect on any Javascript benchmarks.
Unfortunately tracemonky still doesn't work on 64-bit builds. You can set the javascript.options.jit.content preference to true with no warning that you are at most going to get a placebo effect on javascript performance. So anyone with a modern system won't be able to take advantage of one of the biggest new features.
A relational database is not merely a storage engine. This is a common mistake made by MySQL users (and developers).
Days later, the download page still links to 2.4.0 for English PPC Mac. You can get OO 3.0 in Japanese and French though.
I tried it on a Intel Mac and it works pretty well. Didn't find any bugs for a couple days. The first bug was fairly serious though. It gave an bogus error when I tried to save my file. Even Save As was broken. I was finally able to Save As a Word .doc file, re-open, and re-save as the original ODT. Later, after reducing the font size I got a few glitches in the font display, required toggling zoom levels to fix.
But overall much faster than NeoOffice, and much better than earlier attempts on Mac OS X.
On the other hand, if your boss is able to just barely write something with PHP and MySQL, he might make you actually use these worst-in-class choices!
No one is going to to offer HD-DVD compatibility. It is dead, why would any CE company want to reduce profits by including that support? Slashdot's "collective wisdom" about this whole format war has had an abysmal track record. From people claiming there wasn't a significant difference in quality on the new formats, to the bogus claims that Warner and Toshiba were paid off by Sony, and now the bitter HD-DVD fans and Microsoft shills claiming Blu-ray is just as dead and that downloads are the new thing. The silly thing is that with even a small amount of research the outcome of the war was a foregone conclusion.
This is not tied to T-Mobile hotspots.
You can. Just get a phone that supports UMA and your WiFi calls just use your minutes. The $10 a month is optional, to get unlimited WiFi calling.
If I was in Steve Balmer's shoes, I'd take them off.
I like reiserfs because I can trust it to perform well on any file system load. I can put it on a server and know it will be fast and efficient regardless of what the users do. Ext3 gives ext2 journaling, but does not add efficient large directories or small files, two features that reiserfs has.
Sure ext3 may benchmark slightly faster in certain scenarios. But unless you know ahead of time that those are the only scenarios you are going to put on the file system, I recommend reiserfs.
> Are you shpongled?
Yes, oh yes, a thousand times yes!
I'm not sure if I understand you, but you seem to be saying that people who BSDL their code really don't want other people to take advantage of it with a less free license. But this contradicts several of the assertions made by BSDL advocates in this forum that specifically say they don't mind if this happens.
For the record I have absolutely nothing against the BSD license. What irritates me is people who advocate the BSDL as being "more free" on the one hand, and then turn around and go ape-shit when someone wants to include BSD code in a GPL program.
I've actually seen one person suggest that a clause be added to the BSD license that specificly prohibits use in GPL code! Hey pal, you've just reinvented the GPL, moron.
> So, you're not interested in improving someone else's code, not interested in helping the community, not interested in any sort of the
> customary altruistic activities inherent in a gift-culture. Instead, you just want to take.
> The author of a BSD source code can rerelease it at any time under any license he wishes. But because he doesn't, you would take
> license on it against his wishes (not that you legally could). Ungrateful greedy lout!
I simply cannot fathom what's in your confused mind. You lambast the people who choose to take advantage of the very freedom you praise about the BSD license, all in the same breath!
If you were rational you would be _happy_ if someone takes BSD code and relicense it under the GPL. You would say "ah look, this exactly why I support the BSD license. Good work".
(PS. Sorry for the flame tone -- I'm in a bad mood for unrelated reasons)
> If any of the Linux people think they can just >"tack the GPL on" to the file, they are flat >wrong, plain and simple.
No, you are flat wrong, plain and simple. You are also a hypocrit. Your whole argument that FreeBSD is "more free" is that, in addition to the freedoms of the GPL, you also give the freedom to release less free versions. If Microsoft can take BSD code and include it in their less-free licensed software then the GPL people can certainly do the same.
You may be confusing this with actually changing the copyright holder on your code, in which case you are correct, but that is irrelevent to this discussion.
>I be interested in looking at the bug report you filed for that one. You did file one didn't you?
How much does Microsoft pay us for bug reports? They do pay us, don't they?
It was some sort of soccer game. Looked fine, but unfortunately when I asked them they admitted that they are far from having an OpenGL driver _really_ optimized for the (awesome!) fx6+ cards on Linux yet. (IE they haven't even started on DRI support yet.)
Best bet in the _near_ term for Linux is still a 3dfx Voodoo3 card, but hopefully in 6 months we'll be seeing more highend stuff (Wildcat 4000, fx6+, SGIs vpcs, Evans & Sutherland's new thing...) supported.
No. Linux does have OpenGL.
Also, it's not "much more difficult to port applications" because most of them are already using Unix and X, so the port is in fact easy. It's getting the "product" together -- packaging, marketing, support that takes extra effort -- for any new platform.
Again, though, not everyone can hear your internet radio station. Many many people cannot afford even what we think of as "cheap" computers and internet access fees. They can afford a cheap radio.
Yes it is needed. Not everyone can afford a computer and an internet connection, but most can afford a good radio for $15.
Maya is already running on Linux, it's just not released.
They can say I can't pick my nose too, it doesn't make it true. The sooner someone nukes NSI off the face of the planet the better we'll all be.