Shrug; I hardly used suspend on my desktop back when I had it working (no idea whether it still works). Laptops sure you need it (and in fact I run windows on my laptop), but my desktop is half server these days, if I suspended it my website would go down. I don't care so much for OS wars nowadays, but BSD fits my needs (ZFS, apache, customizable desktop transparency) better than anything else I've found.
Anyway, my actual point was that the nvidia binary drivers tend, in my experience, to be better quality than free software video drivers, S3 suspend support being one example of this.
Sounds like you've put together a system that might be worth cleaning up and putting on freshmeat (or similar) for others to use. I'd be interested anyway. Just a thought, if you have the time.
I think you haven't used win95 recently. I installed win98SE on an old box not so long ago, and was shocked to realise the trolls weren't exaggeration: it really does crash, quite literally, every half hour.
Now, I can't remember which ubuntu 11 is, so it's quite possibly one of the completely unusable versions, but any decent modern linux will beat any pre-2k windows easily.
Other adobe free viewers are similarly crap - pdf might have become the format of the web if it weren't for their stupidly bloated viewer, and their SVG thing was terrible the last time I tried it. So I'd look to management not putting any money into their free viewers when they've got the products they're selling for megabucks (photoshop). After all, it doesn't seem like their terrible security record has cost them many users.
There might be a few corrupt politicians, but there are far more who simply see the reality: we need the capacity our nuclear plants provide, and if we haven't built any new ones... guess we have to keep running the old ones, even if it's past the end of their design lifetime. It becomes the lesser of many evils at that point.
Are you suggesting that Joe Engineer who has worked 20 years at JPL is going to pack up his family and leave the sun beaches and smog of Southern California for the smog and human rights oppression of Beijing?
If he thinks that's where he'll get to work on cool stuff then yes, yes yes. JPL types care far more for that than anything else.
Given the general quality level I'd have to disagree; the nvidia drivers work perfectly, even supporting e.g. S3 suspend, and most of the code is the same as their (quite widely tested) windows driver. Wheras with the free intel drivers I still get lots of random display corruption, and those're supposed to be on the good end of free drivers.
The common mortal wouldn't know, but Skype has proprietary encryption that still has not been beaten (Russia even wanted to ban Skype)
Proprietary encryption is worthless, and even in the implausible event that their code does what they think it does, there are still plenty of perfectly good free encryption algorithms out there (AES anyone?)
a really good VoiP codec (revolutionary, really, it was the first real contender for a PC phone).
The codec had nothing to do with that; we all have enough bandwidth to use pretty much anything.
The network and the NAT traversal are what made Skype. Nothing else matters.
I can see you've never tried to write a SIP library. The protocol is a complete nightmare. And even if you get it working, it's got no way of passing through firewalls, which is how Skype got to be popular.
What would be needed is optional lightweight types, so you would have something like float:meters and float:feet and warnings/erros on implicit conversions.
You can get this in java using the JSR308 type-checkers framework
In Britain at least, "pulling the cord" means the emergency-stop mechanism. Signalling for a bus to stop is usually a button; we'd probably just say "ring the bell"
1) Text-only chat (which is bat-guano-insane, IMHO)
Skype handles connecting to the same account from multiple clients far more gracefully than anything else. AIM will boot one of you off. Jabber will route messages to whichever has set the obscure "priority" field higher, or, if it's set to the same for both, a random one (I believe it's even in spec to route different messages for the same conversation to different clients, though I doubt any real server's that crazy). With skype I can just leave my work and home copies signed in, and connect my phone as and when I use it, with no fear of getting random half-conversations in each. That's why I use it.
Re:At the risk of being modded flamebait, etc
on
OpenBSD 4.9 Released
·
· Score: 1
There were legitimate reasons for sun to want the patent provisions in the CDDL. Similar protections were introduced in the GPLv3 - perhaps Sun should have waited for that, but even had ZFS been released under GPLv3 its licensing would still be incompatible with Linux. Did Sun deliberately choose to make ZFS incompatible with Linux? Maybe - but it was only possible because Linux's licensing is equally cranky.
Re:At the risk of being modded flamebait, etc
on
OpenBSD 4.9 Released
·
· Score: 0
The IP is now owned by Oracle, the license isn't truly open-source.
Huh? It's released under the CDDL, which is an open-source license by any reasonable definition; it's approved by the OSI and the FSF, and very similar to the GPLv3 aside from the latter's extended patent licensing provisions. It's incompatible with the Linux kernel because the Linux kernel is GPLed, but that's every bit as much Linux's fault as ZFS's (both use strict copyleft licenses that therefore don't play nice with other licenses).
You're right about Haskell being a beautiful language, but it is not as fast as C/C++.
Depends on the problem. My previous company found the Haskell proxy we wrote for testing could handle 5x the load of the best (thread-based) C++ implementation.
If your argument is worth anything you should be able to outline it; I'm not expecting detail, but just an answer to the simple question "why should I care". I don't believe China becoming more powerful is going to change my life because I simply see no way that it could, and I'm not going to go away and read two books just to satisfy some guy on slashdot.
If you couldn't do pixel perfect CSS layouts then developers would still be using a lot more png's and jpegs for simple things.
And marking themselves out as wrong when they did so, and costing themselves more in bandwidth.
In my ideal world, by default all browsers would render everything exactly the same
That's absurd; think about a phone rendering a page the same as a widescreen monitor. The point of html is that the browser gets to adapt it, reflowing things as it sees fit, rather than having to display everything exactly as the designer positioned it.
You're just repeating the same talk - losing rankings, positions. You mention "flexing muscle", but nothing concrete.
Suppose that in 5 years China is #1 in this ranking. What effect will this have on me, as an ordinary US citizen(*) - how is it going to change my life? Maybe some guy in China will be richer than me, but why do I care? Maybe my company will be owned by Chinese shareholders, but again, why do I care?
(*)I don't actually live in the US, but assume I do
Sadly modern CSS has gone that way too. I cringed when I saw the acid2 test's requirement that the page "look exactly, pixel-for-pixel, like this reference rendering" - it simply shouldn't be possible to specify something like that in html.
Hmm. On one level it is trivial, but it's an immediate thing that gets in the way all the time, and it encourages an excessively verbose style of programming that makes the language worse all round. I don't really miss object literals, and I think the lack of type inference is a deliberate decision; in fact, I'd say explicit typing is what makes java what it is. It could do with closures, but I can usually manage without them; the JSR166 parallel-array libraries let me express most of the map/reduce type logic I usually want to use closures for. I would appreciate more powerful generics (e.g. co- and contravariance), and fewer compiler bugs around the generics that exist, and better type composition (mixins or the like) would be a godsend. But all these are things that annoy me maybe once or twice a week, wheras the getter/setter functions are in my face all the time.
Code is read more often than it's written, you still have to read and understand those methods even though you don't have to write them. Most of the time you end up assuming that getFoo is a dumb accessor for foo and reading straight past it - which will then majorly trip you up the one time in 1000 someone's done something "clever" in the method. Where I work we tried using Lombok for a while as a way around this, but found it introduced compiler bugs.
Which is why we need a better definition of species than the current notion of being able to interbreed - every creature could interbreed with its parents, yet somehow along the way we end up with different species.
Slashdot Mirror
Anyway, my actual point was that the nvidia binary drivers tend, in my experience, to be better quality than free software video drivers, S3 suspend support being one example of this.
Sounds like you've put together a system that might be worth cleaning up and putting on freshmeat (or similar) for others to use. I'd be interested anyway. Just a thought, if you have the time.
Now, I can't remember which ubuntu 11 is, so it's quite possibly one of the completely unusable versions, but any decent modern linux will beat any pre-2k windows easily.
Other adobe free viewers are similarly crap - pdf might have become the format of the web if it weren't for their stupidly bloated viewer, and their SVG thing was terrible the last time I tried it. So I'd look to management not putting any money into their free viewers when they've got the products they're selling for megabucks (photoshop). After all, it doesn't seem like their terrible security record has cost them many users.
There might be a few corrupt politicians, but there are far more who simply see the reality: we need the capacity our nuclear plants provide, and if we haven't built any new ones... guess we have to keep running the old ones, even if it's past the end of their design lifetime. It becomes the lesser of many evils at that point.
The free drivers for my graphics card don't support it, and my sound card modules have to be manually unloaded and reloaded. Maybe I just got unlucky.
Are you suggesting that Joe Engineer who has worked 20 years at JPL is going to pack up his family and leave the sun beaches and smog of Southern California for the smog and human rights oppression of Beijing?
If he thinks that's where he'll get to work on cool stuff then yes, yes yes. JPL types care far more for that than anything else.
Given the general quality level I'd have to disagree; the nvidia drivers work perfectly, even supporting e.g. S3 suspend, and most of the code is the same as their (quite widely tested) windows driver. Wheras with the free intel drivers I still get lots of random display corruption, and those're supposed to be on the good end of free drivers.
The common mortal wouldn't know, but Skype has proprietary encryption that still has not been beaten (Russia even wanted to ban Skype)
Proprietary encryption is worthless, and even in the implausible event that their code does what they think it does, there are still plenty of perfectly good free encryption algorithms out there (AES anyone?)
a really good VoiP codec (revolutionary, really, it was the first real contender for a PC phone).
The codec had nothing to do with that; we all have enough bandwidth to use pretty much anything.
The network and the NAT traversal are what made Skype. Nothing else matters.
I can see you've never tried to write a SIP library. The protocol is a complete nightmare. And even if you get it working, it's got no way of passing through firewalls, which is how Skype got to be popular.
What would be needed is optional lightweight types, so you would have something like float:meters and float:feet and warnings/erros on implicit conversions.
You can get this in java using the JSR308 type-checkers framework
In Britain at least, "pulling the cord" means the emergency-stop mechanism. Signalling for a bus to stop is usually a button; we'd probably just say "ring the bell"
1) Text-only chat (which is bat-guano-insane, IMHO)
Skype handles connecting to the same account from multiple clients far more gracefully than anything else. AIM will boot one of you off. Jabber will route messages to whichever has set the obscure "priority" field higher, or, if it's set to the same for both, a random one (I believe it's even in spec to route different messages for the same conversation to different clients, though I doubt any real server's that crazy). With skype I can just leave my work and home copies signed in, and connect my phone as and when I use it, with no fear of getting random half-conversations in each. That's why I use it.
There were legitimate reasons for sun to want the patent provisions in the CDDL. Similar protections were introduced in the GPLv3 - perhaps Sun should have waited for that, but even had ZFS been released under GPLv3 its licensing would still be incompatible with Linux. Did Sun deliberately choose to make ZFS incompatible with Linux? Maybe - but it was only possible because Linux's licensing is equally cranky.
The IP is now owned by Oracle, the license isn't truly open-source.
Huh? It's released under the CDDL, which is an open-source license by any reasonable definition; it's approved by the OSI and the FSF, and very similar to the GPLv3 aside from the latter's extended patent licensing provisions. It's incompatible with the Linux kernel because the Linux kernel is GPLed, but that's every bit as much Linux's fault as ZFS's (both use strict copyleft licenses that therefore don't play nice with other licenses).
My (AMD Geode-based) Vye S18 is fantastic; sounds like the touchscreen version would do what you want.
You're right about Haskell being a beautiful language, but it is not as fast as C/C++.
Depends on the problem. My previous company found the Haskell proxy we wrote for testing could handle 5x the load of the best (thread-based) C++ implementation.
If your argument is worth anything you should be able to outline it; I'm not expecting detail, but just an answer to the simple question "why should I care". I don't believe China becoming more powerful is going to change my life because I simply see no way that it could, and I'm not going to go away and read two books just to satisfy some guy on slashdot.
If you couldn't do pixel perfect CSS layouts then developers would still be using a lot more png's and jpegs for simple things.
And marking themselves out as wrong when they did so, and costing themselves more in bandwidth.
In my ideal world, by default all browsers would render everything exactly the same
That's absurd; think about a phone rendering a page the same as a widescreen monitor. The point of html is that the browser gets to adapt it, reflowing things as it sees fit, rather than having to display everything exactly as the designer positioned it.
Suppose that in 5 years China is #1 in this ranking. What effect will this have on me, as an ordinary US citizen(*) - how is it going to change my life? Maybe some guy in China will be richer than me, but why do I care? Maybe my company will be owned by Chinese shareholders, but again, why do I care?
(*)I don't actually live in the US, but assume I do
They let "usability experts" design them.
Sadly modern CSS has gone that way too. I cringed when I saw the acid2 test's requirement that the page "look exactly, pixel-for-pixel, like this reference rendering" - it simply shouldn't be possible to specify something like that in html.
Hmm. On one level it is trivial, but it's an immediate thing that gets in the way all the time, and it encourages an excessively verbose style of programming that makes the language worse all round. I don't really miss object literals, and I think the lack of type inference is a deliberate decision; in fact, I'd say explicit typing is what makes java what it is. It could do with closures, but I can usually manage without them; the JSR166 parallel-array libraries let me express most of the map/reduce type logic I usually want to use closures for. I would appreciate more powerful generics (e.g. co- and contravariance), and fewer compiler bugs around the generics that exist, and better type composition (mixins or the like) would be a godsend. But all these are things that annoy me maybe once or twice a week, wheras the getter/setter functions are in my face all the time.
Code is read more often than it's written, you still have to read and understand those methods even though you don't have to write them. Most of the time you end up assuming that getFoo is a dumb accessor for foo and reading straight past it - which will then majorly trip you up the one time in 1000 someone's done something "clever" in the method. Where I work we tried using Lombok for a while as a way around this, but found it introduced compiler bugs.
Which is why we need a better definition of species than the current notion of being able to interbreed - every creature could interbreed with its parents, yet somehow along the way we end up with different species.