How can anybody base his ordinary software on Wine. [...] Any Wine application is still an ordinary Windows application following the Windows design and UI guidelines.
Because the Windows design and UI guidelines aren't actually that bad, and it's not like there's a single consistent UI that people expect their X11 applications to use. Seriously: if they move to native X, should they be using GTK, QT, Motif, or some other (presumably lighter-weight) toolkit? Why not use winelib as an X toolkit? It certainly eases porting concerns.
I've been told that the broadcom-based belkin cards DO work under linux, using the 'ndiswrapper' driver, which basically loads the windows driver. I've not been able to find any to test this with, though, as they seem to have been phased out and replaced with another chipset which isn't compatible with ndiswrapper. Belkin, of course, haven't updated the model number of the cards (F5D7050, IIRC) to reflect the change in chipset, so you can't tell what you're getting until you get the card out of its box.
There's a Linux driver for the new chipset under development, but when I last checked (~5 months ago) it didn't work with my card.
Fair enough. I can only assume you're a libertarian if you think misdescribing goods is not immoral. And you must live somewhere with zero consumer protection laws, so perhaps you have a libertarian legislature?
So set "insughtful" to +5 and all the others to -5, and read the existing discussion at +5...
You know, that would only work if moderators actually moderated accurately. The number of times I've seen an insightful post moderated '+5 funny' or something like that is beyond count. And '+5 insightful' more often than not isn't.
Slashdot's moderation system is great at keeping the trolls under control. It isn't so good at sorting out the good comments from the bad.
Following the law is for users, not Microsoft. They can do what they want.
And I (as a UK citizen, a country in which they have an office) can sue them in the small claims court for recovery of the money I paid for my licence. It'd cost me no more than £60 to issue the claim, and if I lose I might face up to £100 in a costs award. It's worth the risk, because I think a judge wouldn't be overly sympathetic with them.
By EULA you agreed too. NO WARRANTY WRITTEN OR IMPLIED.
In most jurisdictions exclusions of warranties are severely limited in scope by a variety of laws (e.g., in mine, the Unfair Contract Terms Act 1977) and legal precedents. The manufacturer intentionally preventing the product from doing what it was claimed in advertising that it would do is almost certainly not covered by the exclusion.
Unfortunately, you probably can't claim anything more than your money back, even if the sudden unexpected failure cost you significant cash. Unless even the limitation of damages clause is ruled illegal, as it might be in the case of intentional damage.
Doing this would open MS up to a *lot* of legal action. Just one more reason why they won't be doing it.
Requirements:... A directshow compatible DVD player installed
It's a free add-on to an existing DVD player. It can't play DVDs by itself.
(This is an important correction, as I've been looking for a free DVD player for windows other than Media Player Classic for a while, and don't have a directshow DVD player installed)
I don't know what kind of weird hardware you have hooked, up, but My linux box supports all the hardware I throw at it.
I'm not sure about the original poster, but over the years I've had trouble getting the following hardware to work:
* graphics cards produced by Diamond Multimedia (circa 1997-1999). * sound cards described as "Windows Sound System" compatible (circa 1997-1999) * Winbond ISDN cards (circa 2000) * ADSL modems supplied free by British ISPs to their customers (typically Alcatel Speedtouch) (circa 2001) * Belkin wireless ethernet cards that don't have Broadcom chipsets (2004-present)
All of these devices were fairly common at the times I've indicated, and in many cases were practically ubiquitous because they were the cheapest brand available at the time.
I fail to see why this oversimplificaton was egregious to the point of lid-flipping.
Because HT is nothing like as beneficial as "having 2 processors in one" (i.e. a dual-core processor), and as such he was misdescribing the goods in a way that makes them sound much better than they are. I don't know about where you are, but over here that kind of behaviour is illegal. It's certainly immoral.
You could say "HT is like having 2 processors in one, only the second processor is about the same speed as that computer you threw away a couple of years back because it was too slow". That wouldn't be too bad.
Microsoft has learned that to keep their users locked into Microsoft Office formats they have to do things we in the free software world can't do and wouldn't want to do--change the format
Which is presumably why my copy of Office97 isn't able to open documents saved by the latest and greatest versions of MS Office.
Except it can. MS don't do this, and haven't done it since they redesigned the format to use an extendible syntax, which was (I believe) first implemented in Word95. Older versions only fail to open documents that actually use features that didn't exist in the older version. Generally, documents degrade reasonably well to something the older version can show even in that case.
What killed WordPerfect was that they could not open Word Documents
The rumours of WordPerfect's death have been greatly exaggerated. A new version was released only last year. I know a number of people who still use it in their day-to-day work. Most of them are novelists. One is a lawyer. It isn't dead, it has been relegated to niche markets.
The UK system, which allows you to choose what you pay by choosing whether to make calls. Not receiving an incoming call isn't really an option. Holding off on making a call until you can do it from a landline often is.
Yes, I'm aware that in the end, the average customer pays the same. The point is that it is easier to manage your finances with the UK system.
You can't put up with shoddy customer service like this. You're paying them... this means it is up to them to deal with your concerns. If someone hangs up on you, call back. Whoever you get through to, ask to speak to their manager. Tell the manager that you want to complain about somebody who hung up on you. If you're lucky, the employee will have made a note in your file about your call. If you're not, then you may have to persuade the manager to talk to the database management department to get a trace on the last employee who accessed your records before the manager did. They should be willing to do this. If they aren't, go up a level. Eventually, if they continue resisting, you'll end up talking to the call centre manager (it's hard to get higher than this). Such people are usually cooperative.
Threaten to move your account. Threaten to file complaints with any appropriate regulatory bodies. Threaten legal action. Threaten to go to the press. Threats can get you a long way, but always be nice about it. Be reasonable. People will deal with you if you seem like you'll be reasonable.
In UK, if we recieve am abusive call, calling the phone company will not be any help. They will rightly ask you to contact the police first, and they will work with the police to resolve the matter.
Actually, BT has a department that customers can call if they're receiving abusive calls that offers advice about how to deal with them, and actively encourages people to call them before the police. I suspect other phone companies handle it in a similar manner.
The problem is that embedding HTML in RSS is not part of the standard:
The most serious compatibility problem is with HTML markup. Userland's RSS reader--generally considered as the reference implementation--did not originally filter out HTML markup from feeds. As a result, publishers began placing HTML markup into the titles and descriptions of items in their RSS feeds. This behaviour has become widely expected of readers, to the point of becoming a de facto standard, though there is still some inconsistency in how software handles this markup, particularly in titles. (source)
If you aren't making serious advertising money, the bandwidth fees from your amateur video hour would actually run into bankrupting-levels if a blogger got hit with several "instalanches" in one month on top of say, 10,000 regular viewers a month.
If you're getting 10,000 regular viewers per month, you ought to be getting at least 50,000 page hits per month. You can get $2 per 1,000 impressions from advertising almost without lifting a finger. $100 per month ought to pay for some hefty bandwidth. I don't see the problem.
For most companies (yknow, except flickr and textdrive etc), setting up and maintaining a blog is going to have the smallest ROI of any of the approaches you mention, because it will reach only the voracious readers and news junkies of the Internet.
Well, yeah, but ain't the point so obvious that it shouldn't even have been mentioned in a WSJ article. I mean, blogs were designed from the ground up to be a vehicle for personal interaction. Why would anyone think they were useful in marketing?
There are other corporate uses for a blog; allowing non-team members to stay up-to-date on the progress of a particular project, for instance, and feeding back on developments within that project through a comments section. I just don't get why *anyone* would try to use one for marketing. And I've been in the business of selling web-based marketing packages to corporations for the last nine years, so it's not just that its a field I'm unfamiliar with.
Since these temperatures only occurs naturally in space, why not build a super, big cluster of these things, hook them up to a satallite and launch it into orbit.
Because the notion that these temperatures actually occur in space is a total myth, propounded by people with virtually zero physics knowledge trying to talk to people with even less physics knowledge about the inherent problems of space flight.
Temperature is a property of a material, referring to the material's tendency to transfer heat energy into cooler material or away from warmer material due to thermal conductance, and to radiate EM radiation. Space is a (near) vacuum, hence there is no material to have such a property. Therefore, space does not have temperature.
Keeping stuff cool in space is a big problem. If it generates heat (as the transistors on these chips will), then you have to find a way to dump that heat somewhere. Generally, the only option is radiation. Radiation is a very poor way of losing heat if you need to keep your system very cold.
I initially thought that, but then realised that the article doesn't at any point describe what this chip actually does. So, I surmise that it isn't a general purpose processor (which would be a ridiculous leap forward: a processor that clocks in at around 200 times current-gen consumer systems?), but probably a digital signal processor of some kind. 500GHz might then be its sampling frequency, meaning that it could work with 250GHz signals. At this point, comparing its clock speed to the frequency of a radio signal is a useful, meaningful comparison.
Why does Evolution's GUI stand out as much? It doesn't look like a Windows application - the colours are wrong, for one, the toolbar delimiters are non-standard, the up-down widget as well, the checkbox is non-checkboxey, the icons are bland, and there are lots of buttons around.
Because it's a GTK application, and GTK doesn't use native widgets. As others have suggested, you can install a theme to avoid this issue, but it won't fix a few other sticking points that I've had with other GTK apps:
* Non-standard button placement, with cancel on the left and OK on the right (yes, I'm well aware of Apple's research in this area, but it's generally acknowledged that consistency is more important and all other Windows apps do it the other way around) * Bizarre, hard-to-use open file dialog box which doesn't let you type in file names and doesn't recognise shortcuts (so if you double-click on one it tries to open the shortcut file, rather than the file it points to) * Bizarre, hard-to-use save file dialog box which doesn't show a list of files already present in the directory * Placement of application files in a nonstandard place (directly in a subdirectory of the user profile directory, rather than in the Local Settings or Application Data subdirectories) which can screw up the use of roaming profiles
If GTK people want to claim that they support windows, they're going to have to do better IMO. It's just like all the projects that claim to support MacOSX but don't properly integrate with the system. We might as well be running it on a Linux box if it isn't going to work right with the rest of the system.
1) You are playing a PC game and are only getting 10fps. You buy an upgrade for $100 to get you 100 fps. Call this upgrade VHS to DVD. Now you also have the option to spend $1000 and get 500 fps. Call this upgrade DVD to HDDVD (Blu-ray or HD-DVD). Between 10 fps and 100 fps, the user will see a huge difference. However, between 100 fps and 500 fps even though there IS a more significant change, very few people would notice it all.
Bad analogy; there is an absolute limit to the number of frames per second the human eye can take in, and we're pretty close to it at the 50/60 that is common in video playback. Most people can see in a lot more detail than their TV screen can display. The resolution of the human eye is approximately 1 arc minute (although some people have been able to demonstrate significantly better performance). For somebody sitting 2 metres from a 1 metre wide TV display, the horizontal pixels on a DVD-standard TV signal take about 1.4mm each = about 2.4 arc minutes. Ideally, we should therefore have better than 2.4 times DVD resolution (just slightly less than the resolution of 1080p), assuming this is a reasonable TV configuration. If you think 3 metres would be a better viewing distance for that size TV, only 1.6 times the resolution is required (roughly 720p). But, I think you'd be in the minority if you suggested any more distance than that.
what they will notice is when the gc decides it needs to scan a memory area that has been swapped out crowding out any other IO on the system.
Two problems with this statement:
1. An interepreted environment does not necessarily mean that mark & sweep garbage collection is in use. 2. With a generational garbage collector, you don't usually have to scan pages that haven't been modified since the last scan (as these are in old generations and their pointers are already represented in the set of roots for the newer generations), which means you usually only scan stuff that's already in core (except in exceptional low-memory circumstances).
java.io really sucks for some types of apps as it basically forces you to have one thread per socket
Agreed. Java's IO packages are pretty bad; I wasn't really thinking about Java when I wrote this.:)
Absolutely. I've worked on projects using the Java language but targetting a native code environment through an ahead-of-time compiler. It's not the way the language was designed to work, but it can be made to work, and there are advantages to doing it that way (in this case, a lighter-weight distribution package due to not having to include a JRE, and faster start-up times due to lack of JIT overhead).
Agreed. One point worth noting is that I can't think of many ways of producing a dataset that large where the data is produced faster than TrueCrypt can encrypt it. Don't store it on an unencrypted partition and then encrypt it for processing, produce it directly on the encrypted partition and then move the resulting data. Similarly, don't decrypt to local storage at the other end, use the file directly from the encrypted partition; chances are your consuming application (presumably some kind of data analysis or data mining tool) will be slower than the decryption too. You'll want two encrypted partitions so you can alternate which one you're writing to and which one you're moving, but that doesn't need any more space (20GB) than doing the encryption offline would.
Slashdot Mirror
How can anybody base his ordinary software on Wine. [...] Any Wine application is still an ordinary Windows application following the Windows design and UI guidelines.
Because the Windows design and UI guidelines aren't actually that bad, and it's not like there's a single consistent UI that people expect their X11 applications to use. Seriously: if they move to native X, should they be using GTK, QT, Motif, or some other (presumably lighter-weight) toolkit? Why not use winelib as an X toolkit? It certainly eases porting concerns.
I've been told that the broadcom-based belkin cards DO work under linux, using the 'ndiswrapper' driver, which basically loads the windows driver. I've not been able to find any to test this with, though, as they seem to have been phased out and replaced with another chipset which isn't compatible with ndiswrapper. Belkin, of course, haven't updated the model number of the cards (F5D7050, IIRC) to reflect the change in chipset, so you can't tell what you're getting until you get the card out of its box.
There's a Linux driver for the new chipset under development, but when I last checked (~5 months ago) it didn't work with my card.
Fair enough. I can only assume you're a libertarian if you think misdescribing goods is not immoral. And you must live somewhere with zero consumer protection laws, so perhaps you have a libertarian legislature?
So set "insughtful" to +5 and all the others to -5, and read the existing discussion at +5...
You know, that would only work if moderators actually moderated accurately. The number of times I've seen an insightful post moderated '+5 funny' or something like that is beyond count. And '+5 insightful' more often than not isn't.
Slashdot's moderation system is great at keeping the trolls under control. It isn't so good at sorting out the good comments from the bad.
Following the law is for users, not Microsoft. They can do what they want.
And I (as a UK citizen, a country in which they have an office) can sue them in the small claims court for recovery of the money I paid for my licence. It'd cost me no more than £60 to issue the claim, and if I lose I might face up to £100 in a costs award. It's worth the risk, because I think a judge wouldn't be overly sympathetic with them.
By EULA you agreed too. NO WARRANTY WRITTEN OR IMPLIED.
In most jurisdictions exclusions of warranties are severely limited in scope by a variety of laws (e.g., in mine, the Unfair Contract Terms Act 1977) and legal precedents. The manufacturer intentionally preventing the product from doing what it was claimed in advertising that it would do is almost certainly not covered by the exclusion.
Unfortunately, you probably can't claim anything more than your money back, even if the sudden unexpected failure cost you significant cash. Unless even the limitation of damages clause is ruled illegal, as it might be in the case of intentional damage.
Doing this would open MS up to a *lot* of legal action. Just one more reason why they won't be doing it.
Err, no it isn't.
...
Requirements:
A directshow compatible DVD player installed
It's a free add-on to an existing DVD player. It can't play DVDs by itself.
(This is an important correction, as I've been looking for a free DVD player for windows other than Media Player Classic for a while, and don't have a directshow DVD player installed)
I don't know what kind of weird hardware you have hooked, up, but My linux box supports all the hardware I throw at it.
I'm not sure about the original poster, but over the years I've had trouble getting the following hardware to work:
* graphics cards produced by Diamond Multimedia (circa 1997-1999).
* sound cards described as "Windows Sound System" compatible (circa 1997-1999)
* Winbond ISDN cards (circa 2000)
* ADSL modems supplied free by British ISPs to their customers (typically Alcatel Speedtouch) (circa 2001)
* Belkin wireless ethernet cards that don't have Broadcom chipsets (2004-present)
All of these devices were fairly common at the times I've indicated, and in many cases were practically ubiquitous because they were the cheapest brand available at the time.
I fail to see why this oversimplificaton was egregious to the point of lid-flipping.
Because HT is nothing like as beneficial as "having 2 processors in one" (i.e. a dual-core processor), and as such he was misdescribing the goods in a way that makes them sound much better than they are. I don't know about where you are, but over here that kind of behaviour is illegal. It's certainly immoral.
You could say "HT is like having 2 processors in one, only the second processor is about the same speed as that computer you threw away a couple of years back because it was too slow". That wouldn't be too bad.
Microsoft has learned that to keep their users locked into Microsoft Office formats they have to do things we in the free software world can't do and wouldn't want to do--change the format
Which is presumably why my copy of Office97 isn't able to open documents saved by the latest and greatest versions of MS Office.
Except it can. MS don't do this, and haven't done it since they redesigned the format to use an extendible syntax, which was (I believe) first implemented in Word95. Older versions only fail to open documents that actually use features that didn't exist in the older version. Generally, documents degrade reasonably well to something the older version can show even in that case.
What killed WordPerfect was that they could not open Word Documents
The rumours of WordPerfect's death have been greatly exaggerated. A new version was released only last year. I know a number of people who still use it in their day-to-day work. Most of them are novelists. One is a lawyer. It isn't dead, it has been relegated to niche markets.
The UK system, which allows you to choose what you pay by choosing whether to make calls. Not receiving an incoming call isn't really an option. Holding off on making a call until you can do it from a landline often is.
Yes, I'm aware that in the end, the average customer pays the same. The point is that it is easier to manage your finances with the UK system.
Amen.
You can't put up with shoddy customer service like this. You're paying them... this means it is up to them to deal with your concerns. If someone hangs up on you, call back. Whoever you get through to, ask to speak to their manager. Tell the manager that you want to complain about somebody who hung up on you. If you're lucky, the employee will have made a note in your file about your call. If you're not, then you may have to persuade the manager to talk to the database management department to get a trace on the last employee who accessed your records before the manager did. They should be willing to do this. If they aren't, go up a level. Eventually, if they continue resisting, you'll end up talking to the call centre manager (it's hard to get higher than this). Such people are usually cooperative.
Threaten to move your account. Threaten to file complaints with any appropriate regulatory bodies. Threaten legal action. Threaten to go to the press. Threats can get you a long way, but always be nice about it. Be reasonable. People will deal with you if you seem like you'll be reasonable.
In UK, if we recieve am abusive call, calling the phone company will not be any help. They will rightly ask you to contact the police first, and they will work with the police to resolve the matter.
Actually, BT has a department that customers can call if they're receiving abusive calls that offers advice about how to deal with them, and actively encourages people to call them before the police. I suspect other phone companies handle it in a similar manner.
If you aren't making serious advertising money, the bandwidth fees from your amateur video hour would actually run into bankrupting-levels if a blogger got hit with several "instalanches" in one month on top of say, 10,000 regular viewers a month.
If you're getting 10,000 regular viewers per month, you ought to be getting at least 50,000 page hits per month. You can get $2 per 1,000 impressions from advertising almost without lifting a finger. $100 per month ought to pay for some hefty bandwidth. I don't see the problem.
For most companies (yknow, except flickr and textdrive etc), setting up and maintaining a blog is going to have the smallest ROI of any of the approaches you mention, because it will reach only the voracious readers and news junkies of the Internet.
Well, yeah, but ain't the point so obvious that it shouldn't even have been mentioned in a WSJ article. I mean, blogs were designed from the ground up to be a vehicle for personal interaction. Why would anyone think they were useful in marketing?
There are other corporate uses for a blog; allowing non-team members to stay up-to-date on the progress of a particular project, for instance, and feeding back on developments within that project through a comments section. I just don't get why *anyone* would try to use one for marketing. And I've been in the business of selling web-based marketing packages to corporations for the last nine years, so it's not just that its a field I'm unfamiliar with.
Since these temperatures only occurs naturally in space, why not build a super, big cluster of these things, hook them up to a satallite and launch it into orbit.
Because the notion that these temperatures actually occur in space is a total myth, propounded by people with virtually zero physics knowledge trying to talk to people with even less physics knowledge about the inherent problems of space flight.
Temperature is a property of a material, referring to the material's tendency to transfer heat energy into cooler material or away from warmer material due to thermal conductance, and to radiate EM radiation. Space is a (near) vacuum, hence there is no material to have such a property. Therefore, space does not have temperature.
Keeping stuff cool in space is a big problem. If it generates heat (as the transistors on these chips will), then you have to find a way to dump that heat somewhere. Generally, the only option is radiation. Radiation is a very poor way of losing heat if you need to keep your system very cold.
Given the comparison in the article with mobile phone transmission frequencies, I'd guess it was built for some kind of signal processing application.
I initially thought that, but then realised that the article doesn't at any point describe what this chip actually does. So, I surmise that it isn't a general purpose processor (which would be a ridiculous leap forward: a processor that clocks in at around 200 times current-gen consumer systems?), but probably a digital signal processor of some kind. 500GHz might then be its sampling frequency, meaning that it could work with 250GHz signals. At this point, comparing its clock speed to the frequency of a radio signal is a useful, meaningful comparison.
Why does Evolution's GUI stand out as much? It doesn't look like a Windows application - the colours are wrong, for one, the toolbar delimiters are non-standard, the up-down widget as well, the checkbox is non-checkboxey, the icons are bland, and there are lots of buttons around.
Because it's a GTK application, and GTK doesn't use native widgets. As others have suggested, you can install a theme to avoid this issue, but it won't fix a few other sticking points that I've had with other GTK apps:
* Non-standard button placement, with cancel on the left and OK on the right (yes, I'm well aware of Apple's research in this area, but it's generally acknowledged that consistency is more important and all other Windows apps do it the other way around)
* Bizarre, hard-to-use open file dialog box which doesn't let you type in file names and doesn't recognise shortcuts (so if you double-click on one it tries to open the shortcut file, rather than the file it points to)
* Bizarre, hard-to-use save file dialog box which doesn't show a list of files already present in the directory
* Placement of application files in a nonstandard place (directly in a subdirectory of the user profile directory, rather than in the Local Settings or Application Data subdirectories) which can screw up the use of roaming profiles
If GTK people want to claim that they support windows, they're going to have to do better IMO. It's just like all the projects that claim to support MacOSX but don't properly integrate with the system. We might as well be running it on a Linux box if it isn't going to work right with the rest of the system.
1) You are playing a PC game and are only getting 10fps. You buy an upgrade for $100 to get you 100 fps. Call this upgrade VHS to DVD. Now you also have the option to spend $1000 and get 500 fps. Call this upgrade DVD to HDDVD (Blu-ray or HD-DVD). Between 10 fps and 100 fps, the user will see a huge difference. However, between 100 fps and 500 fps even though there IS a more significant change, very few people would notice it all.
Bad analogy; there is an absolute limit to the number of frames per second the human eye can take in, and we're pretty close to it at the 50/60 that is common in video playback. Most people can see in a lot more detail than their TV screen can display. The resolution of the human eye is approximately 1 arc minute (although some people have been able to demonstrate significantly better performance). For somebody sitting 2 metres from a 1 metre wide TV display, the horizontal pixels on a DVD-standard TV signal take about 1.4mm each = about 2.4 arc minutes. Ideally, we should therefore have better than 2.4 times DVD resolution (just slightly less than the resolution of 1080p), assuming this is a reasonable TV configuration. If you think 3 metres would be a better viewing distance for that size TV, only 1.6 times the resolution is required (roughly 720p). But, I think you'd be in the minority if you suggested any more distance than that.
what they will notice is when the gc decides it needs to scan a memory area that has been swapped out crowding out any other IO on the system.
:)
Two problems with this statement:
1. An interepreted environment does not necessarily mean that mark & sweep garbage collection is in use. 2. With a generational garbage collector, you don't usually have to scan pages that haven't been modified since the last scan (as these are in old generations and their pointers are already represented in the set of roots for the newer generations), which means you usually only scan stuff that's already in core (except in exceptional low-memory circumstances).
java.io really sucks for some types of apps as it basically forces you to have one thread per socket
Agreed. Java's IO packages are pretty bad; I wasn't really thinking about Java when I wrote this.
Absolutely. I've worked on projects using the Java language but targetting a native code environment through an ahead-of-time compiler. It's not the way the language was designed to work, but it can be made to work, and there are advantages to doing it that way (in this case, a lighter-weight distribution package due to not having to include a JRE, and faster start-up times due to lack of JIT overhead).
Agreed. One point worth noting is that I can't think of many ways of producing a dataset that large where the data is produced faster than TrueCrypt can encrypt it. Don't store it on an unencrypted partition and then encrypt it for processing, produce it directly on the encrypted partition and then move the resulting data. Similarly, don't decrypt to local storage at the other end, use the file directly from the encrypted partition; chances are your consuming application (presumably some kind of data analysis or data mining tool) will be slower than the decryption too. You'll want two encrypted partitions so you can alternate which one you're writing to and which one you're moving, but that doesn't need any more space (20GB) than doing the encryption offline would.