My friends and other people I talk to will often just reel off names of people who I assume are Hollywood movie stars to me and are often shocked when the names don't mean anything to me. I don't understand how people are willing to maintain a complex mental database of actors and what films they have starred in but are not willing to remember arguably more useful and usually less-complex information.
Often when I see a film I'll recognise an actor and think to myself "oh, that's the guy who was in that film about x", but I don't really know what his name is or anything like that, because I've never really thought of that information as important. I guess that's just me.
A network-transparent windowing system which communicated high-level widget creation/modification messages was my degree dissertation project. One drawback of this approach is that it puts applications at the mercy of the system to support the widgets it needs. This is okay for simple interfaces such as a file browser (just a tree view and an icon list) but something like a wordprocessor would need to draw its main canvas manually.
For the aforementioned project, I designed the server to accept "plugin" widgets and claimed that this would allow application vendors to optionally offload some of their drawing to the display server, but I also implemented a drawing canvas with a simple API, which is in reality how most applications would approach the problem. I toyed with the idea of having a transparent negotiation system where the application would use the server-side implementation if available and otherwise use its own local copy, but that put a lot of extra burden on the application-side which I had to accept that few developers would bother with.
My project is not released, since it is not a complete implementation -- I had to cut it down for the purposes of the project -- and also I'm still waiting for clearance from the university that they are not interested in the copyright on it, so I'm not legally allowed to distribute it yet.
I believe what he meant was that geeks like choices and options. Most usability specialists recommend against having options for everything and instead suggest that the developer just decide which is the "correct" choice. They have a point to a certain extent, but often this is often taken too far by well-meaning developers and it reaches a point where specific classes of users are being obstructed from the software because they don't fit into the "common case" the software was designed for.
Remember that the GPL is a licence to distribute and create derivative works, not a licence for use. If the GPL was found to be invalid, everyone could go on using the software produced by the original author, but would no longer be able to create derived versions or re-distribute it.
There are some hazy problems to do with already-existing redistributions: would all of the packaged Linux distributions become illegal since they contain software distributed without a licence? Perhaps, but I think that it's safe to say that any author who licenced code under the GPL will not file suit against Red Hat and SuSE for copyright infringment, since they obviously intended the software to be distributed in this manner regardless of the specific wording of the licence.
Not to mention the game is then limited to the current generation of Windows systems. There will come a time in the future when Microsoft changes the driver interface again due to massive kernel changes (as they did from 9x to NT). Hopefully Microsoft would retain some back-compat with the old interface, but in that case you'd have to have a "compatability mode" driver running and slowing down your system just to play some old game. Worst case is that Microsoft has to break back-compat completely and the game becomes useless unless people want to downgrade to Windows XP.
Fortunately, past experience with id games suggests that just before their games pass out of active support they'll disable the CD checks in one of the last update patches as they have for Quake 2 and Quake 3. I guess this is the best compromise they can get between keeping the publisher happy and keeping the fans happy. I don't think any of the major players at id are under any delusion that this copy-protection actually helps anything; I'm betting it's mandated by Activision.
Doesn't help with non-id games, of course. I've seen other developers do similar things, but only for the high-profile games which get actively maintained for months or years after release. Many games are just a case of "release and forget".
There are plenty of interesting people to meet online. I've met a whole load of people from different places in various "community"-type sites like the various "blogging" communities and even slashdot. However, I've not met anyone on Friendster or Orkut; there it seems a bit "forced". While on LiveJournal and its clone sites the social networking is just a side-effect, on Orkut it is all there is to it. There is no other reason for it. I stopped logging in a month or so after I signed up just because there was nothing left to do on there but delete all of the community spam.
This LOAF thing isn't really anything to do with social networking, though, having said all that. It seems more like a way of "rating" people who send messages do you by favouring those that are once-removed from your circle of friends. In the long run this might help you meet new people since these new people are of course emailing you something, but that is not really the point of it.
The solution to the cross-platform problem won't be a bunch of class wrappers around native widgets. Instead, that problem will be solved with higher-level conceptual classes which don't have a one-to-one relationship with the underlying widgets. It would be another layer between the business logic and the UI code which takes the high-level interface ideas and makes the right calls to make it happen in a sensible way for each system.
Of course, making a completely general concept-based class library isn't easy. However, I'm sure someone could achieve a compromise which manages to encapsulate many of the common ideas present in today's GUIs. For example, rather than having a Window class which can contain Controls, you could (for suitable classes of application) have a DocumentCollection class which can contain Documents. On Windows this could be rendered as an MDI interface with a child for each document (I know MDI is deprecated, but this is just an example) while on MacOS this would presumably just be a bunch of top-level windows which "stick together" in the window order. This is a pretty poor example, but I'm sure you can think of some of your own -- the higher-level they are the better.
For the purposes of your application, though, you could make your own abstractions which are totally specific to whatever you're writing. The only bit that has to differ between platforms is a a small set of API calls in response to methods from the higher-level API above. Most of the "presentation logic" which actually makes the interface do something would be in the high-level interface layer shared between them all.
I do admit, though, that designing a high-level GUI abstraction layer for many programs would be tedious and unrewarding. For larger projects, though, it'd be a useful approach. I have a small attachment to the methods I've outlined here because my degree dissertation project was in part based around these ideas, although not applied exactly as I'm describing here. I won't claim it's the best approach for everything, though; toolkits like SWT for Java or wxWidgets which abstract at the widget/component level can be very useful in many cases.
It sounds to me like your current approach is how it should be done. Make sure none of your other code depends on the UI and then just write several UI layers. That way it'll look native wherever you run it.
In my opinion, GTk# on Windows should just be used to run software that was never intended to run on Windows, as a compatibility layer in similar ven to Cygwin. If you are actually indending your application to be cross-platform, having several switchable GUI layers is the way to go. Later on, if it seems worthwhile, you can put a Cocoa# GUI layer on it too and you have a Mac version with no change to the guts of your application. You could even make a terminal-based user interface to parts of your core application as appropriate without changing the "interesting" parts at all. If your app is open source, other people might even contribute alternative interfaces for you!
What are you talking about? There are no adverts at all on the BBC. Sure, they have trailers advertising other shows but these are always between programmes, not during them. Unless you're planning to just veg out in front of the TV and consume, you'll probably never see these trailers because you'll turn on the TV to watch whatever you wanted to watch and turn it off again at the end.
I did something similar, except I don't accept dialin to my house. Instead, I just have my phone configured to dial up to a local rate ISP (which are still common in the UK despite the rise of the poorly-implemented unmetered dialup) and use my net-facing IP address as the WAP gateway. There I have kannel running as you mentioned. There are a couple of public WAP gateways out there for if you don't want to run your own, too, but the ones I tried wouldn't deal with my POST requests which apparently require some protocol extension I forget the name of.
Since I have DSL at home this works just fine, and I don't need to deal with people phoning my house and getting greeted by the chirps of a modem. My bare-bones cellphone (a Nokia 3410, I seem to remember) doesn't do GPRS or this fancy WAP 2.0 stuff. I don't use the WAP browser but instead little Java Midlet applications which pull data down by HTTP. That way I can pull down a bunch of data, disconnect and then use it "offline". It only takes a minute or so to sync the data, which includes any email marked as unread in my mailbox and a few other things of interest probably only to me. The efficient retrieval mechanism in conjunction with dialin (by 9600bps ISDN, in fact) to a local rate number makes it pretty cheap. I think the only thing I ever used the standard WAP browser for was retrieving my MIDlets onto the phone in the first place! (no bluetooth on this phone either)
I once attempted to write a Browser Helper Object for IE which would automatically do something like that script. Unfortunately, I started from a position of not having done any COM programming before, so I got buried under a mountain of COM learning and never actually got to the point. I wish someone who was an experienced COM developer would work on it, though. The next step would be to employ the same tricks that get Gator, Bonzai Buddy and all that other junk installed on systems to get the BHO installed so that people who don't know or care what PNG is will still benefit! ;)
Unfortunately, there is no similar hack available for background images. People can make pretty alpha-transparent logos, but not semi-transparent text boxes with a subtle background texture showing through.
If Microsoft has no developers assigned to that project then it isn't relevant to this discussion, since we're talking about why Microsoft would want to improve IE. They obviously don't want to improve MacIE since they abandoned it.
I think that makes my original point quite well. Want the latest IE? You (essentially) need Windows. Some weird geeks have managed to run it under Wine, but I don't think Microsoft has anything to worry about from that point of view for some time yet.
Anyone who uses Internet Explorer is using Windows. If Internet Explorer has the highest market share, they control the de facto standards and can keep customers locked into Internet Explorer and by extension Windows.
This is the attitude throughout Microsoft. It's the same reason why the Windows API still sucks after 20 years. The vast majority of customers don't give a damn about any of this stuff because they don't care: no sites will use alpha-transparent PNG unless IE does, so why bother implementing it?
This is the problem with relying on commercial entities for "innovation": they'll only bother when it actually benefits them. Mozilla, on the other hand, implement things purely for the sake of completeness and interest.
Hey, don't hate on me. It was the parent to my post who started this poor analogy. My analogy was just an example to show that his didn't map properly.
Saying that politics is completely based on opinion is a bit far-fetched, though. Sure, there are a lot of points of view and no obvious correct answer, but there are clearly good and bad approaches and picking the good approach requires at least some background knowledge and experience. Social sciences are a bit more wavy than hard sciences, but there's still lots of expertise involved.
Perhaps a better analogy would be that people with degrees in computer science aren't allowed to do any research to further the field. The grandparent poster was arguing that politicians are making the law overly complex purely to reinforce the need for lawyers to interpret the law. Likewise, computer science researchers attempt to create new fields of computer science (AI following on to machine vision and natural language processing and so forth) which creates a need for new computer scientists to fill these new fields of interest.
Unfortunately, lots of law has to be complex because society is complex. Sure, there have been some "stupid" laws recently, but with your average non-lawyer citizen making laws it'll just get messy, not simple.
> My program crashed. What does that suggest to you? > There is a bug somewhere. Does it make you happy to know there is a bug somewhere? > No. Why are you so negative? > Are you going to help me fix this bug? We were discussing you, not me.
Even ignoring market forces, I still can't see any benefit to using free software vs. unlicenced comercial software.
If (say) 10 people buy Windows XP Professional and 10 people download a CD image from BitTorrent and use a leaked corporate CD key, Microsoft gets the money for ten copies of Windows XP Professional.
On the other hand, if 10 people buy Windows XP Professional and 10 people download the CD image of Debian Woody Binary-1, Microsoft gets the money for ten copies of Windows XP Professional.
How is the second situation any better than the first, bearing in mind the stipulation that we ignore the "Software A gets more mindshare" factor for a moment? As far as Microsoft's concerned, they've seen the same result in both cases. The ten licenced users of Windows XP Professional still use Windows XP Professional in both cases. The public at large doesn't care either way what software these 20 people are using. The only people who see any direct effect here are the ten who are now using Debian instead of Windows XP Professional, and whether that is better obviously depends on the user and doesn't directly affect anyone else.
(This is really just a variation on the "I wouldn't use it if I didn't get it for free" argument for the use of unlicenced software, of course.)
I was a little disappointed when I read what the guy had really done. He's just de-cased his consoles and put them in a big box.
I imagined some clever wiring/electronics to have all the consoles driven by one set of four controllers, with one SCART/Component output for all of the consoles, a cleverer switch setup that doesn't boil down to having one reset button for each console, etc. Now that would have been neat, although
I suppose hardcore Nintendo fans would balk at the idea of using the wrong kind of controller with the wrong game -- a Game Cube controller for the NES? NOOOOOO! I assume, also, that the GC controllers are a bit more complex than just some pushbuttons hooked up to some wires as the old controllers were. Still, it would have been cool!
No-one should be serving XHTML 2.0 directly to browsers except in very specific, controlled circumstances until browsers support it -- which might, of course, be never.
XHTML's more useful in the short term for using as input to XSLT transformations, or as part of other XML Applications such as Atom or SVG. Of course, no-one should be using XHTML 2.0 in that situation either, since it's not a W3C Recommendation yet.
What exactly are the Sith "avenging"? I can't think of anything the "good guys" have done so far that wasn't provoked by the "bad guys", so as far as I can see there's no revenge here -- just business as usual for the bad guys.
Clearly I don't trust the government to spend money wisely. Much of my post was about how badly they are spending money already.
My post was intended to present two slightly-related points. Firstly, voters are taking the lowering of taxes as a good thing without questioning why the system is unable to operate efficiently with the current taxes. Secondly, the reason why the system is unable to operate efficiently is because the money is being spent inefficiently.
My "two choices" to solve the problem were intended to represent the thought process I want the average voter to adopt: If the system sucks now, then either they need to raise taxes or allocate the money better. Lowering taxes clearly isn't the solution. If they care enough to keep thinking, they'll look and find that the NHS and the education system in the UK are wasting money left, right and centre, but even if they don't bother to do any research they can at least see that "raising taxes is a possible solution" doesn't match "Vote for us and we'll lower taxes!"
The obvious argument here is to vote for someone who can spend the money more effectively. If one of the candidates presented a sensible budget allocation plan, and also saved the lowering of taxes thing until after the plan has succeeded, I'd vote for him. Using lower taxes as a marketing gimmick is what I have an issue with, along with the fact that people fall for it.
Here in Britain there was a popular BBC sitcom, filmed documentary-style, set in the office of a Slough paper merchant called Wernam Hogg. Most of the time it's about the boss being a total wad, but one one episode we did encounter the IT staff, which consists of a single technician-type called Simon. Observe:
Tim: Alright Simon? (secretly pretends to decapitate him from behind his back) How's it going?.........
Tim: er... what are you doing with my computer?
Simon: Umm... it's not *your* computer is it? It's Wernam Hogg's.
Tim: Alright, what are you doing with Wernam Hogg's computer?
Simon: You don't need to know.
Tim: No, I don't need to know, but could you tell me anyway?
Simon: *sigh* I'm installing a firewall.
Tim: Okay, What's that?
Simon: It protects your computer against script kiddies, data collectors, viruses, worms *and* trojan horses and it limits your outbound Internet communications. Any more questions?
Tim: Yes. How long will it take?
Simon: Why, do you want to do it yourself?
Tim: No, I can't do it myself. How long will it take you, out of interest?
Simon: It'll take as long as it takes.
Tim: How long did it take last ti.... ?
Simon: It's done.
Tim: Right, thank you.
Simon: Now, I'm going to switch it off. When it comes back on it'll ask you to hit Yes, No or Cancel. Hit Cancel. Do *not* hit Yes or No.
Tim: Right.
Simon: Did you hear what I said?
Tim: Yep.
Simon: What did I say?
Tim: Hit Cancel.
Simon: Good.
Now, obviously most IT staff aren't quite that bad, but the fact that it's been satirised in this manner suggests that is the general opinion of IT staff among non-IT people. This doesn't bode well.
I personally have the experience of replacing a previous freelance IT consultant for a small firm -- too small to have their own in-house IT staff. After working for a few weeks, one of the staff commented that they are a lot more relaxed around me than they were around the old guy, who apparently was very condescending and acted, apparently, like Simon above. I don't really consider myself to have *excellent* social skills, but I get by. I find it distasteful that society has such a low opinion of IT staff in general because of people like my predecessor and Simon from The Office. Why is it that the same stereotypes aren't applied to any other profession?
Slashdot Mirror
My friends and other people I talk to will often just reel off names of people who I assume are Hollywood movie stars to me and are often shocked when the names don't mean anything to me. I don't understand how people are willing to maintain a complex mental database of actors and what films they have starred in but are not willing to remember arguably more useful and usually less-complex information.
Often when I see a film I'll recognise an actor and think to myself "oh, that's the guy who was in that film about x", but I don't really know what his name is or anything like that, because I've never really thought of that information as important. I guess that's just me.
A network-transparent windowing system which communicated high-level widget creation/modification messages was my degree dissertation project. One drawback of this approach is that it puts applications at the mercy of the system to support the widgets it needs. This is okay for simple interfaces such as a file browser (just a tree view and an icon list) but something like a wordprocessor would need to draw its main canvas manually.
For the aforementioned project, I designed the server to accept "plugin" widgets and claimed that this would allow application vendors to optionally offload some of their drawing to the display server, but I also implemented a drawing canvas with a simple API, which is in reality how most applications would approach the problem. I toyed with the idea of having a transparent negotiation system where the application would use the server-side implementation if available and otherwise use its own local copy, but that put a lot of extra burden on the application-side which I had to accept that few developers would bother with.
My project is not released, since it is not a complete implementation -- I had to cut it down for the purposes of the project -- and also I'm still waiting for clearance from the university that they are not interested in the copyright on it, so I'm not legally allowed to distribute it yet.
I believe what he meant was that geeks like choices and options. Most usability specialists recommend against having options for everything and instead suggest that the developer just decide which is the "correct" choice. They have a point to a certain extent, but often this is often taken too far by well-meaning developers and it reaches a point where specific classes of users are being obstructed from the software because they don't fit into the "common case" the software was designed for.
Remember that the GPL is a licence to distribute and create derivative works, not a licence for use. If the GPL was found to be invalid, everyone could go on using the software produced by the original author, but would no longer be able to create derived versions or re-distribute it.
There are some hazy problems to do with already-existing redistributions: would all of the packaged Linux distributions become illegal since they contain software distributed without a licence? Perhaps, but I think that it's safe to say that any author who licenced code under the GPL will not file suit against Red Hat and SuSE for copyright infringment, since they obviously intended the software to be distributed in this manner regardless of the specific wording of the licence.
IANAL, of course.
Not to mention the game is then limited to the current generation of Windows systems. There will come a time in the future when Microsoft changes the driver interface again due to massive kernel changes (as they did from 9x to NT). Hopefully Microsoft would retain some back-compat with the old interface, but in that case you'd have to have a "compatability mode" driver running and slowing down your system just to play some old game. Worst case is that Microsoft has to break back-compat completely and the game becomes useless unless people want to downgrade to Windows XP.
Fortunately, past experience with id games suggests that just before their games pass out of active support they'll disable the CD checks in one of the last update patches as they have for Quake 2 and Quake 3. I guess this is the best compromise they can get between keeping the publisher happy and keeping the fans happy. I don't think any of the major players at id are under any delusion that this copy-protection actually helps anything; I'm betting it's mandated by Activision.
Doesn't help with non-id games, of course. I've seen other developers do similar things, but only for the high-profile games which get actively maintained for months or years after release. Many games are just a case of "release and forget".
There are plenty of interesting people to meet online. I've met a whole load of people from different places in various "community"-type sites like the various "blogging" communities and even slashdot. However, I've not met anyone on Friendster or Orkut; there it seems a bit "forced". While on LiveJournal and its clone sites the social networking is just a side-effect, on Orkut it is all there is to it. There is no other reason for it. I stopped logging in a month or so after I signed up just because there was nothing left to do on there but delete all of the community spam.
This LOAF thing isn't really anything to do with social networking, though, having said all that. It seems more like a way of "rating" people who send messages do you by favouring those that are once-removed from your circle of friends. In the long run this might help you meet new people since these new people are of course emailing you something, but that is not really the point of it.
Maybe your grocer means "Fresh Fish is For Sale"! ;)
The solution to the cross-platform problem won't be a bunch of class wrappers around native widgets. Instead, that problem will be solved with higher-level conceptual classes which don't have a one-to-one relationship with the underlying widgets. It would be another layer between the business logic and the UI code which takes the high-level interface ideas and makes the right calls to make it happen in a sensible way for each system.
Of course, making a completely general concept-based class library isn't easy. However, I'm sure someone could achieve a compromise which manages to encapsulate many of the common ideas present in today's GUIs. For example, rather than having a Window class which can contain Controls, you could (for suitable classes of application) have a DocumentCollection class which can contain Documents. On Windows this could be rendered as an MDI interface with a child for each document (I know MDI is deprecated, but this is just an example) while on MacOS this would presumably just be a bunch of top-level windows which "stick together" in the window order. This is a pretty poor example, but I'm sure you can think of some of your own -- the higher-level they are the better.
For the purposes of your application, though, you could make your own abstractions which are totally specific to whatever you're writing. The only bit that has to differ between platforms is a a small set of API calls in response to methods from the higher-level API above. Most of the "presentation logic" which actually makes the interface do something would be in the high-level interface layer shared between them all.
I do admit, though, that designing a high-level GUI abstraction layer for many programs would be tedious and unrewarding. For larger projects, though, it'd be a useful approach. I have a small attachment to the methods I've outlined here because my degree dissertation project was in part based around these ideas, although not applied exactly as I'm describing here. I won't claim it's the best approach for everything, though; toolkits like SWT for Java or wxWidgets which abstract at the widget/component level can be very useful in many cases.
It sounds to me like your current approach is how it should be done. Make sure none of your other code depends on the UI and then just write several UI layers. That way it'll look native wherever you run it.
In my opinion, GTk# on Windows should just be used to run software that was never intended to run on Windows, as a compatibility layer in similar ven to Cygwin. If you are actually indending your application to be cross-platform, having several switchable GUI layers is the way to go. Later on, if it seems worthwhile, you can put a Cocoa# GUI layer on it too and you have a Mac version with no change to the guts of your application. You could even make a terminal-based user interface to parts of your core application as appropriate without changing the "interesting" parts at all. If your app is open source, other people might even contribute alternative interfaces for you!
What are you talking about? There are no adverts at all on the BBC. Sure, they have trailers advertising other shows but these are always between programmes, not during them. Unless you're planning to just veg out in front of the TV and consume, you'll probably never see these trailers because you'll turn on the TV to watch whatever you wanted to watch and turn it off again at the end.
For more about this theory, see Why Are There So Few Female Computer Scientists? by Ellen Spertus.
I did something similar, except I don't accept dialin to my house. Instead, I just have my phone configured to dial up to a local rate ISP (which are still common in the UK despite the rise of the poorly-implemented unmetered dialup) and use my net-facing IP address as the WAP gateway. There I have kannel running as you mentioned. There are a couple of public WAP gateways out there for if you don't want to run your own, too, but the ones I tried wouldn't deal with my POST requests which apparently require some protocol extension I forget the name of.
Since I have DSL at home this works just fine, and I don't need to deal with people phoning my house and getting greeted by the chirps of a modem. My bare-bones cellphone (a Nokia 3410, I seem to remember) doesn't do GPRS or this fancy WAP 2.0 stuff. I don't use the WAP browser but instead little Java Midlet applications which pull data down by HTTP. That way I can pull down a bunch of data, disconnect and then use it "offline". It only takes a minute or so to sync the data, which includes any email marked as unread in my mailbox and a few other things of interest probably only to me. The efficient retrieval mechanism in conjunction with dialin (by 9600bps ISDN, in fact) to a local rate number makes it pretty cheap. I think the only thing I ever used the standard WAP browser for was retrieving my MIDlets onto the phone in the first place! (no bluetooth on this phone either)
I once attempted to write a Browser Helper Object for IE which would automatically do something like that script. Unfortunately, I started from a position of not having done any COM programming before, so I got buried under a mountain of COM learning and never actually got to the point. I wish someone who was an experienced COM developer would work on it, though. The next step would be to employ the same tricks that get Gator, Bonzai Buddy and all that other junk installed on systems to get the BHO installed so that people who don't know or care what PNG is will still benefit! ;)
Unfortunately, there is no similar hack available for background images. People can make pretty alpha-transparent logos, but not semi-transparent text boxes with a subtle background texture showing through.
If Microsoft has no developers assigned to that project then it isn't relevant to this discussion, since we're talking about why Microsoft would want to improve IE. They obviously don't want to improve MacIE since they abandoned it.
I think that makes my original point quite well. Want the latest IE? You (essentially) need Windows. Some weird geeks have managed to run it under Wine, but I don't think Microsoft has anything to worry about from that point of view for some time yet.
Anyone who uses Internet Explorer is using Windows. If Internet Explorer has the highest market share, they control the de facto standards and can keep customers locked into Internet Explorer and by extension Windows.
This is the attitude throughout Microsoft. It's the same reason why the Windows API still sucks after 20 years. The vast majority of customers don't give a damn about any of this stuff because they don't care: no sites will use alpha-transparent PNG unless IE does, so why bother implementing it?
This is the problem with relying on commercial entities for "innovation": they'll only bother when it actually benefits them. Mozilla, on the other hand, implement things purely for the sake of completeness and interest.
Hey, don't hate on me. It was the parent to my post who started this poor analogy. My analogy was just an example to show that his didn't map properly.
Saying that politics is completely based on opinion is a bit far-fetched, though. Sure, there are a lot of points of view and no obvious correct answer, but there are clearly good and bad approaches and picking the good approach requires at least some background knowledge and experience. Social sciences are a bit more wavy than hard sciences, but there's still lots of expertise involved.
Perhaps a better analogy would be that people with degrees in computer science aren't allowed to do any research to further the field. The grandparent poster was arguing that politicians are making the law overly complex purely to reinforce the need for lawyers to interpret the law. Likewise, computer science researchers attempt to create new fields of computer science (AI following on to machine vision and natural language processing and so forth) which creates a need for new computer scientists to fill these new fields of interest.
Unfortunately, lots of law has to be complex because society is complex. Sure, there have been some "stupid" laws recently, but with your average non-lawyer citizen making laws it'll just get messy, not simple.
> My program crashed.
What does that suggest to you?
> There is a bug somewhere.
Does it make you happy to know there is a bug somewhere?
> No.
Why are you so negative?
> Are you going to help me fix this bug?
We were discussing you, not me.
Even ignoring market forces, I still can't see any benefit to using free software vs. unlicenced comercial software.
If (say) 10 people buy Windows XP Professional and 10 people download a CD image from BitTorrent and use a leaked corporate CD key, Microsoft gets the money for ten copies of Windows XP Professional.
On the other hand, if 10 people buy Windows XP Professional and 10 people download the CD image of Debian Woody Binary-1, Microsoft gets the money for ten copies of Windows XP Professional.
How is the second situation any better than the first, bearing in mind the stipulation that we ignore the "Software A gets more mindshare" factor for a moment? As far as Microsoft's concerned, they've seen the same result in both cases. The ten licenced users of Windows XP Professional still use Windows XP Professional in both cases. The public at large doesn't care either way what software these 20 people are using. The only people who see any direct effect here are the ten who are now using Debian instead of Windows XP Professional, and whether that is better obviously depends on the user and doesn't directly affect anyone else.
(This is really just a variation on the "I wouldn't use it if I didn't get it for free" argument for the use of unlicenced software, of course.)
I was a little disappointed when I read what the guy had really done. He's just de-cased his consoles and put them in a big box.
I imagined some clever wiring/electronics to have all the consoles driven by one set of four controllers, with one SCART/Component output for all of the consoles, a cleverer switch setup that doesn't boil down to having one reset button for each console, etc. Now that would have been neat, although
I suppose hardcore Nintendo fans would balk at the idea of using the wrong kind of controller with the wrong game -- a Game Cube controller for the NES? NOOOOOO! I assume, also, that the GC controllers are a bit more complex than just some pushbuttons hooked up to some wires as the old controllers were. Still, it would have been cool!
No-one should be serving XHTML 2.0 directly to browsers except in very specific, controlled circumstances until browsers support it -- which might, of course, be never.
XHTML's more useful in the short term for using as input to XSLT transformations, or as part of other XML Applications such as Atom or SVG. Of course, no-one should be using XHTML 2.0 in that situation either, since it's not a W3C Recommendation yet.
What exactly are the Sith "avenging"? I can't think of anything the "good guys" have done so far that wasn't provoked by the "bad guys", so as far as I can see there's no revenge here -- just business as usual for the bad guys.
Clearly I don't trust the government to spend money wisely. Much of my post was about how badly they are spending money already.
My post was intended to present two slightly-related points. Firstly, voters are taking the lowering of taxes as a good thing without questioning why the system is unable to operate efficiently with the current taxes. Secondly, the reason why the system is unable to operate efficiently is because the money is being spent inefficiently.
My "two choices" to solve the problem were intended to represent the thought process I want the average voter to adopt: If the system sucks now, then either they need to raise taxes or allocate the money better. Lowering taxes clearly isn't the solution. If they care enough to keep thinking, they'll look and find that the NHS and the education system in the UK are wasting money left, right and centre, but even if they don't bother to do any research they can at least see that "raising taxes is a possible solution" doesn't match "Vote for us and we'll lower taxes!"
The obvious argument here is to vote for someone who can spend the money more effectively. If one of the candidates presented a sensible budget allocation plan, and also saved the lowering of taxes thing until after the plan has succeeded, I'd vote for him. Using lower taxes as a marketing gimmick is what I have an issue with, along with the fact that people fall for it.
Here in Britain there was a popular BBC sitcom, filmed documentary-style, set in the office of a Slough paper merchant called Wernam Hogg. Most of the time it's about the boss being a total wad, but one one episode we did encounter the IT staff, which consists of a single technician-type called Simon. Observe:
.........
Tim: Alright Simon? (secretly pretends to decapitate him from behind his back) How's it going?
Tim: er... what are you doing with my computer?
Simon: Umm... it's not *your* computer is it? It's Wernam Hogg's.
Tim: Alright, what are you doing with Wernam Hogg's computer?
Simon: You don't need to know.
Tim: No, I don't need to know, but could you tell me anyway?
Simon: *sigh* I'm installing a firewall.
Tim: Okay, What's that?
Simon: It protects your computer against script kiddies, data collectors, viruses, worms *and* trojan horses and it limits your outbound Internet communications. Any more questions?
Tim: Yes. How long will it take?
Simon: Why, do you want to do it yourself?
Tim: No, I can't do it myself. How long will it take you, out of interest?
Simon: It'll take as long as it takes.
Tim: How long did it take last ti.... ?
Simon: It's done.
Tim: Right, thank you.
Simon: Now, I'm going to switch it off. When it comes back on it'll ask you to hit Yes, No or Cancel. Hit Cancel. Do *not* hit Yes or No.
Tim: Right.
Simon: Did you hear what I said?
Tim: Yep.
Simon: What did I say?
Tim: Hit Cancel.
Simon: Good.
Now, obviously most IT staff aren't quite that bad, but the fact that it's been satirised in this manner suggests that is the general opinion of IT staff among non-IT people. This doesn't bode well.
I personally have the experience of replacing a previous freelance IT consultant for a small firm -- too small to have their own in-house IT staff. After working for a few weeks, one of the staff commented that they are a lot more relaxed around me than they were around the old guy, who apparently was very condescending and acted, apparently, like Simon above. I don't really consider myself to have *excellent* social skills, but I get by. I find it distasteful that society has such a low opinion of IT staff in general because of people like my predecessor and Simon from The Office. Why is it that the same stereotypes aren't applied to any other profession?