I'd love to do that, but getting a package into Debian is a nightmare that I have simply given up on. Even the simple guides are 50 pages long and a mass of not quite up-to-date information.
Ubuntu makes it trivial. Even if you can't or don't want to get into Ubuntu base, you can just make a PPA on Launchpad and get automatic building for all supported editions and archs of Ubuntu.
We have something like that at VISL, but with zero statistical or machine learning or AI aspects.
We instead write a few thousand rules by hand (largest language has 10000 rules) that look at the context - where context is the entire sentence, and possibly previous or next sentences - to figure out what meaning of a word is being used and what it attaches to.
This kind of system takes longer to develop and refine, but it also doesn't have any of the statistical problems. 95-99% "understanding" of text? Sure, we can do that. Statistics top out long before, and then have to add in rules to get the last 5-10%. And where statistics require giga- or terabytes of text, rule based systems only require a single example of a valid grammatical construct or word usage.
jEdit's only flaw is that it is Java. Asides from that, it is a solid Unicode capable cross-platform editor that can work with files over SSH. Synchronizing your sessions, configuration, and plugins is as simple as copying over your.jedit folder, even between Windows and Linux.
I so far haven't found a single other editor with all those features. Do tell me if one exists...
It does? I've can't see any ads in Skype anywhere, but then I don't use the Home screen...you can turn that thing off. I did enable it just now to see if there are ads, and there are not. It looks like a rather plain contact's status updates listing.
and wants me to issue facebook updates of some shit.
It does? Again, you can turn that off, or just ignore it.
But really, Compact View fixes most ills. It makes Skype look like it always has.
It yields info such as: black magic, Orlando magic, ceremonial magic... magic kingdom, magic roundabout, magic flute... practice magic, radiate magic... magic of animation... etc
At my job (plug: http://gramtrans.com/ ), we're doing human level machine translation now, if by human you're talking about high school level of quality.
But in difference to Google or Microsoft or all the previous attempts, we're NOT doing Statistical Machine Translation; we make Rule-Based MT.
Google's way is to scan huge amounts of parallel texts for good translation candidates, which works to some degree, but it has an upper limit that no amount of parallel text will help them surpass.
Our way involves writing a huge amount of rules so that our software actually understands the input text, and then rules for how things should be translated. This takes a lot longer to develop as you might imagine, but the results are great and there is no limit to how good it gets...if something isn't quite right, add a rule for it.
However, because it takes so long to develop and that we're only 2 people working on it, we only have a few really good translation chains.
- since IE is the only browser to enforce XHR caching, every request needs a timestamp query parameter (something that no other browser does, and which is really stupid, altough easy to provide)
Or, you could use POST requests like you're supposed to. Any number of intermediary caches and load balancing tricks can play foul with GET requests, but POST is not allowed to be cached.
With jQuery this is as simple as using $.post() instead of $.get().
Seriously? You're paranoid about letting go of your card for the 3 seconds it takes to enter the PIN? The card remains right in front of you, no more than 4cm away from your hands...
Where do you live where stealing cards at the payment terminal in full public view is so frequent that you feel a need to be paranoid about it? I've never even heard of such a case of theft/assault.
No, the real problem with the chip system is that when you put the card in the holder, the security code is facing away from you, visible to the store clerk...
Denmark already has a similar thing. We can perform most actions dealing with the government online, and we even get a gratis certificate for digital signing and encryption of emails. I haven't had to go to a government or city office in years.
I have looked around. I've tried them all, pretty much, on both Windows and Linux. Only one I haven't tried is XCode on OS X, but I've heard good things about that. I've dearly wanted to find a single powerful IDE that I could use cross-platform, but none exists.
Of all the IDEs I've tried on various platforms, none come even close to VC++ Express. A huge amount of that rests in the debugger alone...gdb and IDEs that use gdb are hilariously underpowered compared to VC++'s integrated debugger.
So for all development and debugging, I use VC++. For profiling and sanity checking, I use gprof/gcov/valgrind. Best tools for their job.
I'd even say Visual Studio is the best, by a very long mile. There just isn't anything as powerful in a single cohesive package. It is one of the main reasons I develop primarily on Windows (the other being gaming). And don't let anyone fool you into thinking that VC++ is evil...it doesn't lock you into making Windows only programs; you have to lock yourself in, if that's your desire.
You could get similar results with cobbling together various tools, but why bother when VC++ Express already Just Works...
That said, the final pin to make VC++ Express perfect would be some Valgrind functionality.
They don't want to do that. They're even constraining SC2 to have a limited field of vision where regardless of screen resolution you can't zoom out too much.
Personally, I hate that and want to be able to see everything at once, but they feel that players should be limited in this fashion. I doubt you'd see the original SC updated to allow zoom out for the same reason.
Rebate with contract=$100 Rebates are illegal here. Companies are not allowed to use rebates as it is considered tricking the customer.
Companies are well aware that a lot of people will forget to send in rebates, and even if they do there are insane procedures to get a "correct" rebate fulfilled.
So, prices are probably the same after adjusting for tactics. In the US, companies plan prices according to x% getting rebates. In the EU, companies plan flat prices.
Which makes me wonder...how does the accountants and stock peddlers like the idea of rebates. Must throw some nice chaotic numbers into the books. Imagine if everyone (100%) suddenly decided to make good on rebates...
A great deal of what needs doing can be done with standard key combos plus a few links ("shortcut" files) in a folder that's in your %PATH% environment variable. If you enjoy that setup, you should take a look at Launchy (http://www.launchy.net/). Just hit Alt+Space and type in some letters from the program you want to run. It'll match against anything in your Start menu.
Why don't people respect WC3 like they do SC? Personally, WC3 just didn't support battles in the same scale as SC's epic battles. WC3 had 90 "food" where single units take anywhere from 1 to 8 of those. SC had 200 "food" (or 600 if you played Protoss - Mind Control other races' workers).
Even when making custom maps in WC3, raising the food or hero limit wasn't possible. You had to upset the balance of the game by changing units' requirements to get more units in the game.
The only thing that nags me about Postgres is the stupid dump+restore upgrading cycle...
With MySQL all I have to do is service stop, rpm -Uvh, service start, and it all works.
With Postgres I have to disable all access to ensure what I dump is the latest data, then do a dump, then service stop, then rm -rf the data dir, then rpm -Uvh, then service start, then restore, then enable access again.
Even so, I use Postgres. Just wish it was easier to upgrade.
Judging from screenshots and info: the ability to control audio volume per application from the OS. It's as if every app now has its own dedicated Master channel that the OS then mixes together for output.
I've been wanting that for years due to certain apps that think they are divine and simply take over/mute my global Master/Wave channel when they feel like it (AIM and Winamp, I'm looking at you!). In Windows Vista, those intolerant apps will not be able to take over.
Lazy app writers who simply use the global channels instead of opening a per-app one should be shot...but with Vista they can continue to be lazy.
Just don't let anyone know exactly where you are running that. 'auto_prepend_file' can be altered at.htaccess level, so if that's the first they change the rest is moot.
I've always thought registering to vote is silly. In order for a democracy to work, you need as many as possible to vote, so requiring them to register first is inconvenient (and we all want convenience).
Here in Denmark, every person over 18 is sent a card and a place/time to vote. No registration. The result is that 80+% actually vote...
Yeah ok, so we are a small (5.3 million) country so it's easier to manage here. Still makes it a much better way.
"Think about it.. without extensions you'd have to open every file (well, the OS would) in order to tell you what it was.. all formats would have to contain information on their content too - in a standard way (is that image a jpg or a gif?)
This would be really slow."
Windows, curse its name, does this even with extensions. And yes it's slow as hell...
Just try to have a folder full of MPEG/AVI, HTML or EXE files and scroll down through it...it will read each and every file for extra data to show in the status bar, or for the embedded icon.
What I want is a file system, much like Mac OS 9-, with meta data in a seperate fork. Or simply to be able to set the MIME type of a file, regardless of what name it has.
Of course, this is incompatible with the rest of the world, just as Mac OS's system was. But it's still a much better design. Kinda like the x86 vs PPC...the new and improved exists, but we still cling to the old. Blame it on human nature, I guess.
-/-
But that's all regarding actual file systems. When it comes to the web, the 'file' becomes vague. Instead we have the URI/URL scheme, which does not need an extension or even a name. URLs such as http://domain.tld/movie/ should still play if served with a MIME type of a movie, but there's no filename. What will XP SP2 do in such situations, and why should it do anything except deal with the contents based on MIME type?
I'm sure there are plenty of other examples, but in all cases enforcing arbitrary extensions onto the web is not the solution.
Slashdot Mirror
I'd love to do that, but getting a package into Debian is a nightmare that I have simply given up on. Even the simple guides are 50 pages long and a mass of not quite up-to-date information.
Ubuntu makes it trivial. Even if you can't or don't want to get into Ubuntu base, you can just make a PPA on Launchpad and get automatic building for all supported editions and archs of Ubuntu.
We have something like that at VISL, but with zero statistical or machine learning or AI aspects.
We instead write a few thousand rules by hand (largest language has 10000 rules) that look at the context - where context is the entire sentence, and possibly previous or next sentences - to figure out what meaning of a word is being used and what it attaches to.
E.g.
Input: "They're looking at writing an AI which can in some sense understand what is being said."
Output: http://dl.dropbox.com/u/62647212/visl-eng.txt , http://dl.dropbox.com/u/62647212/visl-eng.png
This kind of system takes longer to develop and refine, but it also doesn't have any of the statistical problems. 95-99% "understanding" of text? Sure, we can do that. Statistics top out long before, and then have to add in rules to get the last 5-10%. And where statistics require giga- or terabytes of text, rule based systems only require a single example of a valid grammatical construct or word usage.
Is leaving the child screaming for an hour in the bassinet extreme neglect...
Why yes, yes it is: http://google.com/search?q=cry+it+out+brain+damage
jEdit's only flaw is that it is Java. Asides from that, it is a solid Unicode capable cross-platform editor that can work with files over SSH. Synchronizing your sessions, configuration, and plugins is as simple as copying over your .jedit folder, even between Windows and Linux.
I so far haven't found a single other editor with all those features. Do tell me if one exists...
The current version has no compact buddy list
Sure it does: View -> Compact View
requires a subscription for multiparty video
True, that sucks.
has giant ads on a useless "home" screen
It does? I've can't see any ads in Skype anywhere, but then I don't use the Home screen...you can turn that thing off.
I did enable it just now to see if there are ads, and there are not. It looks like a rather plain contact's status updates listing.
and wants me to issue facebook updates of some shit.
It does? Again, you can turn that off, or just ignore it.
But really, Compact View fixes most ills. It makes Skype look like it always has.
Then maybe what you want is DeepDict. E.g., magic is used like http://gramtrans.com/deepdict/lookup.php?word=magic&class=N&lang=eng&top=200 - it is not free, though all words starting with 's' are currently open to viewing for anyone.
It yields info such as: black magic, Orlando magic, ceremonial magic ... magic kingdom, magic roundabout, magic flute ... practice magic, radiate magic ... magic of animation ... etc
(disclaimer: I work on the DeepDict project)
At my job (plug: http://gramtrans.com/ ), we're doing human level machine translation now, if by human you're talking about high school level of quality.
But in difference to Google or Microsoft or all the previous attempts, we're NOT doing Statistical Machine Translation; we make Rule-Based MT.
Google's way is to scan huge amounts of parallel texts for good translation candidates, which works to some degree, but it has an upper limit that no amount of parallel text will help them surpass.
Our way involves writing a huge amount of rules so that our software actually understands the input text, and then rules for how things should be translated. This takes a lot longer to develop as you might imagine, but the results are great and there is no limit to how good it gets...if something isn't quite right, add a rule for it.
However, because it takes so long to develop and that we're only 2 people working on it, we only have a few really good translation chains.
Agreed. I also have Reply All as default, for the same reasons.
It's just easier, plus it makes you more aware in the rare cases when you don't want to Reply All.
- since IE is the only browser to enforce XHR caching, every request needs a timestamp query parameter (something that no other browser does, and which is really stupid, altough easy to provide)
Or, you could use POST requests like you're supposed to. Any number of intermediary caches and load balancing tricks can play foul with GET requests, but POST is not allowed to be cached.
With jQuery this is as simple as using $.post() instead of $.get().
Seriously? You're paranoid about letting go of your card for the 3 seconds it takes to enter the PIN? The card remains right in front of you, no more than 4cm away from your hands...
Where do you live where stealing cards at the payment terminal in full public view is so frequent that you feel a need to be paranoid about it? I've never even heard of such a case of theft/assault.
No, the real problem with the chip system is that when you put the card in the holder, the security code is facing away from you, visible to the store clerk...
Denmark already has a similar thing. We can perform most actions dealing with the government online, and we even get a gratis certificate for digital signing and encryption of emails. I haven't had to go to a government or city office in years.
Tabs on top is horrible...one of the key reasons I dislike Google Chrome.
I have looked around. I've tried them all, pretty much, on both Windows and Linux. Only one I haven't tried is XCode on OS X, but I've heard good things about that. I've dearly wanted to find a single powerful IDE that I could use cross-platform, but none exists.
Of all the IDEs I've tried on various platforms, none come even close to VC++ Express. A huge amount of that rests in the debugger alone...gdb and IDEs that use gdb are hilariously underpowered compared to VC++'s integrated debugger.
So for all development and debugging, I use VC++. For profiling and sanity checking, I use gprof/gcov/valgrind. Best tools for their job.
I'd even say Visual Studio is the best, by a very long mile. There just isn't anything as powerful in a single cohesive package. It is one of the main reasons I develop primarily on Windows (the other being gaming). And don't let anyone fool you into thinking that VC++ is evil...it doesn't lock you into making Windows only programs; you have to lock yourself in, if that's your desire.
You could get similar results with cobbling together various tools, but why bother when VC++ Express already Just Works...
That said, the final pin to make VC++ Express perfect would be some Valgrind functionality.
They don't want to do that. They're even constraining SC2 to have a limited field of vision where regardless of screen resolution you can't zoom out too much.
Personally, I hate that and want to be able to see everything at once, but they feel that players should be limited in this fashion. I doubt you'd see the original SC updated to allow zoom out for the same reason.
Companies are well aware that a lot of people will forget to send in rebates, and even if they do there are insane procedures to get a "correct" rebate fulfilled.
So, prices are probably the same after adjusting for tactics. In the US, companies plan prices according to x% getting rebates. In the EU, companies plan flat prices.
Which makes me wonder...how does the accountants and stock peddlers like the idea of rebates. Must throw some nice chaotic numbers into the books. Imagine if everyone (100%) suddenly decided to make good on rebates...
(and it's open source: http://sourceforge.net/projects/launchy/ )
Even when making custom maps in WC3, raising the food or hero limit wasn't possible. You had to upset the balance of the game by changing units' requirements to get more units in the game.
WC3 was simply too small. SC was epic.
The only thing that nags me about Postgres is the stupid dump+restore upgrading cycle...
With MySQL all I have to do is service stop, rpm -Uvh, service start, and it all works.
With Postgres I have to disable all access to ensure what I dump is the latest data, then do a dump, then service stop, then rm -rf the data dir, then rpm -Uvh, then service start, then restore, then enable access again.
Even so, I use Postgres. Just wish it was easier to upgrade.
Judging from screenshots and info: the ability to control audio volume per application from the OS. It's as if every app now has its own dedicated Master channel that the OS then mixes together for output.
I've been wanting that for years due to certain apps that think they are divine and simply take over/mute my global Master/Wave channel when they feel like it (AIM and Winamp, I'm looking at you!). In Windows Vista, those intolerant apps will not be able to take over.
Lazy app writers who simply use the global channels instead of opening a per-app one should be shot...but with Vista they can continue to be lazy.
Just don't let anyone know exactly where you are running that. 'auto_prepend_file' can be altered at .htaccess level, so if that's the first they change the rest is moot.
...and in replying to myself, I just read the little snippet saying 'shelved'. Oh well.
Ask and you shall receive: http://imdb.com/title/tt0367882/
I've always thought registering to vote is silly. In order for a democracy to work, you need as many as possible to vote, so requiring them to register first is inconvenient (and we all want convenience).
Here in Denmark, every person over 18 is sent a card and a place/time to vote. No registration. The result is that 80+% actually vote...
Yeah ok, so we are a small (5.3 million) country so it's easier to manage here. Still makes it a much better way.
"Think about it.. without extensions you'd have to open every file (well, the OS would) in order to tell you what it was.. all formats would have to contain information on their content too - in a standard way (is that image a jpg or a gif?)
This would be really slow."
Windows, curse its name, does this even with extensions. And yes it's slow as hell...
Just try to have a folder full of MPEG/AVI, HTML or EXE files and scroll down through it...it will read each and every file for extra data to show in the status bar, or for the embedded icon.
What I want is a file system, much like Mac OS 9-, with meta data in a seperate fork. Or simply to be able to set the MIME type of a file, regardless of what name it has.
Of course, this is incompatible with the rest of the world, just as Mac OS's system was. But it's still a much better design. Kinda like the x86 vs PPC...the new and improved exists, but we still cling to the old. Blame it on human nature, I guess.
-/-
But that's all regarding actual file systems. When it comes to the web, the 'file' becomes vague. Instead we have the URI/URL scheme, which does not need an extension or even a name. URLs such as http://domain.tld/movie/ should still play if served with a MIME type of a movie, but there's no filename. What will XP SP2 do in such situations, and why should it do anything except deal with the contents based on MIME type?
I'm sure there are plenty of other examples, but in all cases enforcing arbitrary extensions onto the web is not the solution.