"The solution is to get a motherboard which properly integrates the SATA controller into the system chipset."
I believe (I could be wrong) that this has more to do with the "mode" of the SATA interface. If the SATA interface is in "legacy emulation mode" (or something like that), the hardware presents the SATA ports to the OS as traditional ATA devices (primary/secondary, master/slave). This limits available features (I'm fuzzy on the details) but maintains the best compatability. In "native" mode (there's some four-leter abbrev I forget), all the legacy IDE/ATA baggage is jettisoned and the OS sees the SATA ports for what they are, but the OS also has to have knowledge of and drivers for same to work.
I know on the Dell OptiPlex's we get, we don't need drivers for the one mode, but do for the other.
"Don't you mean the ARM? Analog Rights Management?"
All this hashing about Digital vs Whatever Rights Management has been a horrible case of missing the point. It isn't the Digital part that matters. The part that matters is that I paid good money for a book (some "content" in buzz-speak) but other people are telling me when and how I should read it.
That's the point. Stop arguing over the freaking "D".
"it seriously gave me the impression that this Longhorn thing was nothing more than a candy shell slapped on top of the same shit MS has been selling for years."
It appears we have someone trying to use this subthread as an advertising opportunity for "Dependency Walker". Aside from the parent post, look here and here. I don't know if this guy is a paid shill or just an over-enthusiastic fan-boy, but clearly, there is a pattern here. People would be well advised to be aware of this, and perhaps moderators should act.
This has been a public service message of the Listmaster General.
"No way in hell can anyone use RegMon from my experience with it... The windows registry is constantly being written to so fast..."
RegMon has filters. You tell it to filter on only the one application you've having trouble with. You start the RegMon capture, start the problem program, wait for the trouble, stop the capture. Then you use filters and search to find the problem. You only need to show the errors -- successful accesses are not problems and can be ignored.
"What you need is Dependency Walker."
How, exactly, does that help me find out what registry and filesystem locations a poorly-written program is trying to access? For that matter, how does it help me fix a program that doesn't want to run unless it has admin rights?
(No, I don't really expect a reply from an AC, but I wanted to make these points in case anyone else is reading this forum for good info.)
The problem with Wikipedia performance isn't so much for readers (although, certainly, fast page loads are desired) but contributors. Wiki's depend on people adding content, editing things, discussing things, following links and doing random fixups and all sorts of other things like that. Sometimes, Wikipedia's performance gets so bogged that this becomes near-impossible. So yah, it's a real problem, and one the admin's are constantly having to deal with.
"Running windows without admin rights is a nightmare."
It certainly isn't easy, unless you're willing to invest significant technical time and effort into the project -- which is, I'm sure, a big part of the reason why most people don't do it.
That being said, I'm the admin for an organization with about 60 or so Windoze stations, and I can say that it can be done for most things. It most often involves figuring out what the defective program is trying to do, and then allowing it access to just where it needs.
The two most vital tools are FileMon and RegMon, both free from SysInternals (http://www.sysinternals.com/). They monitor file system or registry accesses. In the vast majority of programs can be made to work just by applying some ACLs on program-specific registry or filesystem branches.
There's no way in hell your "typical home user" could do this, though, which is, I expect, the problem and point.
"These are very special tools only supplied to special electrical appliances shops, and usually not available within handreach from your chair..."
If they were readily available, I wouldn't have asked. I haven't seen anything like that in any of the stores I frequent. Those include CompUSA, BestBuy, Circuit City, Staples, and OfficeMax. I also haven't seen them at either the local PC sales shows or the local hamfest. Of course, I didn't know such a thing existed (although it seems obvious now), so I wasn't asking after it in particular. But none of the displays had anything like it. I live in the North-East USA, if that matters.
I also have tried some Google seaches, and the only one I was able to find was this one http://www.hammondmfg.com/1580.htm, which doesn't even have TVSS.
(The above is a near-repeat of this post, but the parent and the mod's apparently think I'm just a lazy bastard. While I may be a lazy bastard, I'm not just a lazy bastard.:) I am seriously unable to find anything like the strip in the picture referenced in my original post.
"Are you serious? Just about every office/computer/electrical store here sells boards like that."
Of course I'm serious. If they were readily available, I wouldn't have asked. I haven't seen anything like that in any of the stores I frequent. Those include CompUSA, BestBuy, Circuit City, Staples, and OfficeMax. I also haven't seen them at either the local PC sales shows or the local hamfest. Of course, I didn't know such a thing existed (although it seems obvious now), so I wasn't asking after it in particular. But none of the displays had anything like it. I live in the North-East USA, if that matters.
Hey, they have this neat power strip with each outlet has its own switch. You can see it real well in the upper-left corner of the third picture ("testbed.jpg"). Where do I get one of those? It would be real useful for lab testing (like Asus is doing, duh). The closest I've seen are those under-monitor jobs, but those take up too much room and can't be wall-mounted. What I see here could be.
From TFA: "[Corporations] feel that the changing consumer attitude towards cookies is harming cookie usefulness and unfairly lumping them with spyware and viruses."
And I feel that the changing corporate attitude towards consumers is harming corporate usefulness and unfairly lumps consumers in with goods and resources.
It appears that more and more businesses are starting to see customers as almost like a parasite -- something that drains away their profit margins. They seem to have lost sight of the fact that those same people who are calling and complaining and insisting that stuff actually provide value are the same people who bought their product in the first place. Businesses do not exist in spite of the customer -- the exist because of the customer. Seems obvious, yet increasingly often, I find I need to point this out to "Customer Service Reps" and the like. I find I need to use phrases like, "Look, I'm the flipping customer here!" This is not a sign of a healthy relationship.
"We are not seats or eyeballs or end users or consumers. We are human beings -- and our reach exceeds your grasp. Deal with it." -- cluetrain.org
It's basically the same thing as your standard in-window unit, except that rather then putting the whole thing in the window, you just put a couple of flexible air ducts (hoses) in the window, and put the unit on the floor. You still get the compressor noise, of course, but you do get a good amount of cooling without needing any special installation. They cost a bit more then an in-window unit, but not too too much. I've used them to cool computer rooms before.
"The Windows 'emulation' in OS/2 was never good, and you'd recognize this if you'd ever spent any time dealing with it."
I still own a licensed copy of OS/2 Warp 3 "Blue Spine". I could dig it out of my closet if I wanted to. And the Windows subsystem worked great. It ran Windows programs better then native Windows did. But you have to remember: We're talking about Windows 3.x here.
Almost all of the stuff you list came much later, in Windows 95, 98, etc.
TrueType came with Windows 98, I think. As a patch to 95 at the earliest. DirectX was never supported on Win 3.x. Themes didn't appear until Win98. I dunno what you mean by "expansions". "Fun" is subjective.
I can't argue with the ease-of-use thing. But again, consider that Win 3.x running on top of MS-DOS 5.x was the contemporary of OS/2 Warp, and things don't look nearly as bad.
OS/2 "died" as a serious alternative way before the modern Internet revolution that gave us pirate music, mainstream porn, Shockwave/Flash, and all the other crap.
Those are all facts. So I'm afraid your reasoning is bogus from the get-go. Sorry.
Now, in my opinion, OS/2 died for a lot of reasons. Microsoft's OEM lock-in, IBM's inept marketing, IBM/Microsoft legal fighting, IBM not investing sufficient effort in OS/2 to make it a viable consumer platform, bad luck... lots of factors combined. No one big thing. Again, my opinion.
"It's just a matter of issuing "apt-get dist-upgrade" on the console..."
For individuals who don't have anything major to loose or anything special to worry about, sure. But not for large organizations with a support structure (help desk, local docs, procedures, etc.) that needs to be ramped up to support new changes. And not for anyone doing anything special or mission-critical that needs to test things before deployment. The rule in any production environment is "Test, test, test, and then test some more". You simply cannot just type "apt-get dist-upgrade" (or "yum upgrade" or any other variation on the theme) in the Real World, I'm afraid.
In general, I find that this whole concept (which is a major part of the disipline called "configuration management") appears to be alien to Debian people. When your business/mission is on the line, answers like "Just pull from sid" or "Just apt-get the fix" and so on just don't cut it.
"Debian is offering the "must work"... alternative, and its useful for someone to perform the heavy testing and fixing they do."
While I think that's a good goal to have, and I aplaud Debian for shooting for it, the amount of time it is taking to get this release out is in itself hurting that goal. The world's hardware has been changing rapidly while Debian has been sitting still. Getting a strict Woody config to install on modern hardware can be near-impossible. That is bad.
(And please don't come back with "Just pull from sid"; that rather defeats the purpose of having a stable, tested release without rapid changes and with a consistent configuration management profile. If I wanted that, I'd just run Fedora Core.)
To answer your question: Pascal is still used for teaching in some cultures. Delphi (and the Object Pascal language it provides) still have a following in some corporate circles, especially for database front-end work. And there's a fair bit of legacy code out there.
Trivia: The original Macintosh System Software (later renamed MacOS, later renamed MacOS Classic) was written mainly in Pascal, with assembler where needed for speed or low-level implementation.
I doubt much new, interesting work is done in Pascal, though.
Pascal had some real advantages over it's contemporaries -- K&R C, BASIC, and other things now even more forgotten then Pascal. It was easy to parse, which made a Pascal compiler fast to run and easy to write/maintain. The syntax used more English words and less punctuation, which is arguably easier on the newbie.
Pascal has lots of redundancies and checks, both in the syntax and in the runtime, which made it a lot easier to write and maintain a good, robust program. Some call this "B&D programming"; others, myself included, call it common sense. I don't expect to crash my car, but I still wear a seatbelt. I use my turn signal even when I don't think there is a car in the next lane. I'm human. I make mistakes. I try to make sure the damage from my mistakes is limited.
Pascal also encouraged good programming practices in an era where there was still debate over whether good programming practices really mattered. It popularized the idea of teaching structured programming from the start (as opposed to in a footnote on page 378 of the textbook).
While the original Pascal specification made it of limited use for "real world" stuff, adaptations (like Borland's venerable Turbo Pascal) gave you all the power of C or even assembly when you needed it.
These days, most of the lessons that Pascal taught have been learned, and learned well in some new languages. Many of the things learned in the creation and growth of Pascal have also been learned, leading to languages which are all-around better. New ideas (like OO) have taken hold. Better hardware makes things like garbage collection and runtime evaluation a lot more practical.
So the need for Pascal itself, in the present day, is pretty minimal. However, it played a critical role in the evolution of computer programming as a science and as a professional discipline. It was the "first real language" many people learned. And much like a classic car that's been eclipsed by more modern technology, Turbo Pascal still has a certain elegance and appeal to those who knew it. Nostalgia, yes, but good stuff, still.
I'll try and answer some of your questions, or at least shed some light on them. A lot of this really comes down to "historical reasons" -- it was created that way 30+ years ago, and so we're still stuck with it today. Kinda like the "creat()" function in C/Unix.:-)
"...randomly placing components around a window makes it hard to group and line up things..."
I think they expect you to use the alignment tools to fix that up. Like you say, Windoze background. The idea of having software arrange your widgets/controls for you is too foreign.
"...every expression is terminated with a semi-colon, like C, except for the last one in a code-block, which is optiona."
Not quite. In C, semicolons are, indeed, statement terminators. In Pascal, they are statement separators. That's why you see the behavior you do. For better or worse.
Like you, I took to putting semicolons at the end of most things. I solved the IF problem by using BEGIN/END blocks nearly everywhere. It can be argued that is the right way to go in the long term anyway. Remember, Pascal is designed to encourage good programing practices, and sometimes that increases the short term effort required. Sure, newer languages like Python do a better job, but building Python on the hardware of 30 years ago wouldn't be practical.
"Furthermore, blocks start with 'begin', and end with 'end'. That's alot of characters to type... "
That's why God invented macros.:)
"Finally, a unit is split up in sections like 'interface', 'implementation'."
Turbo Pascal (the ancestor to Delphi and Object Pascal) created units as a way to easily define libraries. You created an "interface", which was the published API for the library -- kinda like a C header file. The "implementation" was the code (like the.c file for a.h file). It provided a form of encapsulation. If you were distributing a unit, you could distribute just the "interface" part and others could still use the unit.
"Why aren't these simply blocks?"
Mainly because they function at a higher level then the normal lexical scopes that BEGIN/END define. In particular, you can define globals that are part of the implementation only, or are also published in the interface.
"And why is the unit itself some sort of half block terminated with 'end.'"
A Pascal program begins with "PROGRAM Foo" and ends with "END."; the Unit syntax just follows suit. No BEGIN was used for the global scope. I expect it's mainly because the "PROGRAM" (or "UNIT") implies you are starting; it also means BEGIN/END are only used to create lexical scopes. The period at the end just signifies the end of the program, same as with an English sentence. It fits Pascal's general approach of trying to provide redundency for safety.
"They just randomly lock up or reboot at various times. These are office machines, not gaming rigs. The one that is giving me fits right now is a fresh install of XP Pro with updates, not a bit of spyware."
Well, something must be wrong with your setup, because I sure don't have that problem. As much as I hate to say it, when properly administered, on reliable hardware, the Windows XP OS itself is pretty stable.
My job includes the care and feeding of about 50 Windoze workstations. The only time I've seen those kinds of problems has been due to either a hardware fault or bad layered software that was installed. The OS itself is stable.
Of course, this is a strongly managed environment. Users don't have any admin or power rights to their computers; we filter anything that even looks like it might be a virus on before it gets near the users; we filter email and web access; web access is through an authenticated proxy server; we test software before it is installed; etc, etc. We do our homework and then some. We also only buy unexciting computers from trustworthy name-brand vendors; no whiteboxes. Because of that, the OS is stable.
Applications still suck. MS Office seems to find a new way to drive me crazy every day, and our ERP system is clunkier then a cement truck on a race track. But the OS itself is stable and confines the damage to individual applications. For what that's worth.
This ain't cheap. Windows costs way more to do right then other OSes, no doubt about it. But it can be done.
While the OP does sound a bit trollish ("Google who can't figure out... even as good a search as their competitors..."), there are some valid points to be had here.
The reason I (and many others, I'm sure) switched from Alta Vista to Google was that Google consistently returned the results I wanted when given a simple query. AV would generally require me to build a relatively complicated expression (often in several iterations) to get good results. Google just needed a word or two. Google's idea of using links to rate page's value ("PageRank") was a brilliant innovation.
Google is still on track in many ways. You'll notice a lot of their recent features can be described as "figuring out what you want", rather then you *telling* Google want you want. That is, if you enter a street address, it gives you a map. If you enter a person's name and town, it gives you their phone book entry. This is a subtle but very powerful and innovative idea. Most systems make you tell the computer what you want. Even if it's a well-designed UI, you're still picking things from drop-down lists and what have you. Google just has a free-form text-entry field, and figures it out. No syntax. No UI other then the English language. Brilliant.
A lot of Google is also just using the web really well. Look at their recent maps system. Maps have been done before by just about everybody, but nobody had come close to doing it as well as Google. That's not so much innovation as just knowing how to wring every ounce of power out of the popular web browsers and JavaScript, but it still counts as progress.
(Geez, I'm starting to sound like a Microsoft press release with all this "innovation" crap. But I believe it's actually true in this case.)
However, despite all this sycophanthy, I think that Google does have serious room for improvement. For one, the OP is right that search engine spam is starting to seriously diminish the value of Google. That's largely a consequence of success. Just like all the viruses target Windoze because that's where the targets are, SEO's target Google. However, Google still needs to do something about it, or Google will continue to loose value. (Same with Microsoft and the virus problem, incidentally.)
Most of the spam on Google is very obvious. It all looks the same, even in the subject lines. One of the things Google does so well is recognize the same kinds of patterns that humans do. So why can't they introduce some heuristics to filter out sites that are selling things? Make it an option, even, so that people can choose. That will even make a lot of the SEO's happier, as they 'll get people who are interest in buying stuff, and not hits from people who don't want their crap.
The search engine spam problem is doubly-bad with Google because of the fact that their UI is so simple. With Altavista, it might take a more complex query, but at least I had the option. With Google, I often lack the tools (query syntax) needed to refine my search and filter out the spam. Lately I've found myself wishing for the power of Altavista's syntax combined with the human-language intelligence of Google.
I don't want Google to become the Novell of the search engine world. Novell, you may recall, had an innovative, unique, and well-done product (NetWare) when they first started to succeed back in the early 1990s. And the early 1990s is right where Novell and NetWare stayed for at least a decade, while everyone else caught up and then passed them.
It would be a shame for the same thing to happen to Google.
Slashdot Mirror
"The solution is to get a motherboard which properly integrates the SATA controller into the system chipset."
I believe (I could be wrong) that this has more to do with the "mode" of the SATA interface. If the SATA interface is in "legacy emulation mode" (or something like that), the hardware presents the SATA ports to the OS as traditional ATA devices (primary/secondary, master/slave). This limits available features (I'm fuzzy on the details) but maintains the best compatability. In "native" mode (there's some four-leter abbrev I forget), all the legacy IDE/ATA baggage is jettisoned and the OS sees the SATA ports for what they are, but the OS also has to have knowledge of and drivers for same to work.
I know on the Dell OptiPlex's we get, we don't need drivers for the one mode, but do for the other.
"news out of /. and the Reg."
:-)
News? Out of Slashdot and The Register? Come on. You might as well include The National Enquirer in that list.
"I might be wrong ( I probably am ) but once its moving doesn't the shuttle go faster than sound in order to achive escape velocity?"
:-)
Faster then sound? Just a bit.
Speed of sound = 760 MPH
Shuttle's orbital velocity = 17,000 MPH
If there's one thing space is good for, it's really impressive numbers.
Granted, I don't know how fast the shuttle would be going while it was in the same altitude band that seagulls fly in.
(All figures are deliberate oversimplications.)
Just some stupid software? Damn! I was hoping to find some episodes of the new Doctor Who for download... ;-)
"Don't you mean the ARM? Analog Rights Management?"
All this hashing about Digital vs Whatever Rights Management has been a horrible case of missing the point. It isn't the Digital part that matters. The part that matters is that I paid good money for a book (some "content" in buzz-speak) but other people are telling me when and how I should read it.
That's the point. Stop arguing over the freaking "D".
"it seriously gave me the impression that this Longhorn thing was nothing more than a candy shell slapped on top of the same shit MS has been selling for years."
;-)
You must be new here.
It appears we have someone trying to use this subthread as an advertising opportunity for "Dependency Walker". Aside from the parent post, look here and here. I don't know if this guy is a paid shill or just an over-enthusiastic fan-boy, but clearly, there is a pattern here. People would be well advised to be aware of this, and perhaps moderators should act.
This has been a public service message of the Listmaster General.
"No way in hell can anyone use RegMon from my experience with it... The windows registry is constantly being written to so fast..."
RegMon has filters. You tell it to filter on only the one application you've having trouble with. You start the RegMon capture, start the problem program, wait for the trouble, stop the capture. Then you use filters and search to find the problem. You only need to show the errors -- successful accesses are not problems and can be ignored.
"What you need is Dependency Walker."
How, exactly, does that help me find out what registry and filesystem locations a poorly-written program is trying to access? For that matter, how does it help me fix a program that doesn't want to run unless it has admin rights?
(No, I don't really expect a reply from an AC, but I wanted to make these points in case anyone else is reading this forum for good info.)
The problem with Wikipedia performance isn't so much for readers (although, certainly, fast page loads are desired) but contributors. Wiki's depend on people adding content, editing things, discussing things, following links and doing random fixups and all sorts of other things like that. Sometimes, Wikipedia's performance gets so bogged that this becomes near-impossible. So yah, it's a real problem, and one the admin's are constantly having to deal with.
HAH! 300 baud acoustic coupler! ;-)
"Running windows without admin rights is a nightmare."
It certainly isn't easy, unless you're willing to invest significant technical time and effort into the project -- which is, I'm sure, a big part of the reason why most people don't do it.
That being said, I'm the admin for an organization with about 60 or so Windoze stations, and I can say that it can be done for most things. It most often involves figuring out what the defective program is trying to do, and then allowing it access to just where it needs.
The two most vital tools are FileMon and RegMon, both free from SysInternals (http://www.sysinternals.com/). They monitor file system or registry accesses. In the vast majority of programs can be made to work just by applying some ACLs on program-specific registry or filesystem branches.
There's no way in hell your "typical home user" could do this, though, which is, I expect, the problem and point.
"These are very special tools only supplied to special electrical appliances shops, and usually not available within handreach from your chair..."
:) I am seriously unable to find anything like the strip in the picture referenced in my original post.
If they were readily available, I wouldn't have asked. I haven't seen anything like that in any of the stores I frequent. Those include CompUSA, BestBuy, Circuit City, Staples, and OfficeMax. I also haven't seen them at either the local PC sales shows or the local hamfest. Of course, I didn't know such a thing existed (although it seems obvious now), so I wasn't asking after it in particular. But none of the displays had anything like it. I live in the North-East USA, if that matters.
I also have tried some Google seaches, and the only one I was able to find was this one http://www.hammondmfg.com/1580.htm, which doesn't even have TVSS.
(The above is a near-repeat of this post, but the parent and the mod's apparently think I'm just a lazy bastard. While I may be a lazy bastard, I'm not just a lazy bastard.
"Are you serious? Just about every office/computer/electrical store here sells boards like that."
Of course I'm serious. If they were readily available, I wouldn't have asked. I haven't seen anything like that in any of the stores I frequent. Those include CompUSA, BestBuy, Circuit City, Staples, and OfficeMax. I also haven't seen them at either the local PC sales shows or the local hamfest. Of course, I didn't know such a thing existed (although it seems obvious now), so I wasn't asking after it in particular. But none of the displays had anything like it. I live in the North-East USA, if that matters.
Hey, they have this neat power strip with each outlet has its own switch. You can see it real well in the upper-left corner of the third picture ("testbed.jpg"). Where do I get one of those? It would be real useful for lab testing (like Asus is doing, duh). The closest I've seen are those under-monitor jobs, but those take up too much room and can't be wall-mounted. What I see here could be.
From TFA: "[Corporations] feel that the changing consumer attitude towards cookies is harming cookie usefulness and unfairly lumping them with spyware and viruses."
And I feel that the changing corporate attitude towards consumers is harming corporate usefulness and unfairly lumps consumers in with goods and resources.
It appears that more and more businesses are starting to see customers as almost like a parasite -- something that drains away their profit margins. They seem to have lost sight of the fact that those same people who are calling and complaining and insisting that stuff actually provide value are the same people who bought their product in the first place. Businesses do not exist in spite of the customer -- the exist because of the customer. Seems obvious, yet increasingly often, I find I need to point this out to "Customer Service Reps" and the like. I find I need to use phrases like, "Look, I'm the flipping customer here!" This is not a sign of a healthy relationship.
"We are not seats or eyeballs or end users or consumers. We are human beings -- and our reach exceeds your grasp. Deal with it." -- cluetrain.org
"You're a behemoth corporation with more money than I can imagine."
To quote Han Solo, "I dunno... I can imagine an awful lot."
"The MUCH better systems are 'split' ... But... this was Australia... I've NEVER seen a model like that here in the US."
m ay_port_dual.htm
There's things like this:
http://www.fedders.com/catalog/appliances/portac/
It's basically the same thing as your standard in-window unit, except that rather then putting the whole thing in the window, you just put a couple of flexible air ducts (hoses) in the window, and put the unit on the floor. You still get the compressor noise, of course, but you do get a good amount of cooling without needing any special installation. They cost a bit more then an in-window unit, but not too too much. I've used them to cool computer rooms before.
"The Windows 'emulation' in OS/2 was never good, and you'd recognize this if you'd ever spent any time dealing with it."
I still own a licensed copy of OS/2 Warp 3 "Blue Spine". I could dig it out of my closet if I wanted to. And the Windows subsystem worked great. It ran Windows programs better then native Windows did. But you have to remember: We're talking about Windows 3.x here.
Almost all of the stuff you list came much later, in Windows 95, 98, etc.
TrueType came with Windows 98, I think. As a patch to 95 at the earliest. DirectX was never supported on Win 3.x. Themes didn't appear until Win98. I dunno what you mean by "expansions". "Fun" is subjective.
I can't argue with the ease-of-use thing. But again, consider that Win 3.x running on top of MS-DOS 5.x was the contemporary of OS/2 Warp, and things don't look nearly as bad.
OS/2 "died" as a serious alternative way before the modern Internet revolution that gave us pirate music, mainstream porn, Shockwave/Flash, and all the other crap.
Those are all facts. So I'm afraid your reasoning is bogus from the get-go. Sorry.
Now, in my opinion, OS/2 died for a lot of reasons. Microsoft's OEM lock-in, IBM's inept marketing, IBM/Microsoft legal fighting, IBM not investing sufficient effort in OS/2 to make it a viable consumer platform, bad luck... lots of factors combined. No one big thing. Again, my opinion.
I've seen a longer version (don't know where it came from) of the classic Calvin and Hobbes quote:
"It's not the verbing that weirds language so much, but rather, the renounification."
"It's just a matter of issuing "apt-get dist-upgrade" on the console..."
For individuals who don't have anything major to loose or anything special to worry about, sure. But not for large organizations with a support structure (help desk, local docs, procedures, etc.) that needs to be ramped up to support new changes. And not for anyone doing anything special or mission-critical that needs to test things before deployment. The rule in any production environment is "Test, test, test, and then test some more". You simply cannot just type "apt-get dist-upgrade" (or "yum upgrade" or any other variation on the theme) in the Real World, I'm afraid.
In general, I find that this whole concept (which is a major part of the disipline called "configuration management") appears to be alien to Debian people. When your business/mission is on the line, answers like "Just pull from sid" or "Just apt-get the fix" and so on just don't cut it.
"Debian is offering the "must work" ... alternative, and its useful for someone to perform the heavy testing and fixing they do."
While I think that's a good goal to have, and I aplaud Debian for shooting for it, the amount of time it is taking to get this release out is in itself hurting that goal. The world's hardware has been changing rapidly while Debian has been sitting still. Getting a strict Woody config to install on modern hardware can be near-impossible. That is bad.
(And please don't come back with "Just pull from sid"; that rather defeats the purpose of having a stable, tested release without rapid changes and with a consistent configuration management profile. If I wanted that, I'd just run Fedora Core.)
To answer your question: Pascal is still used for teaching in some cultures. Delphi (and the Object Pascal language it provides) still have a following in some corporate circles, especially for database front-end work. And there's a fair bit of legacy code out there.
Trivia: The original Macintosh System Software (later renamed MacOS, later renamed MacOS Classic) was written mainly in Pascal, with assembler where needed for speed or low-level implementation.
I doubt much new, interesting work is done in Pascal, though.
Pascal had some real advantages over it's contemporaries -- K&R C, BASIC, and other things now even more forgotten then Pascal. It was easy to parse, which made a Pascal compiler fast to run and easy to write/maintain. The syntax used more English words and less punctuation, which is arguably easier on the newbie.
Pascal has lots of redundancies and checks, both in the syntax and in the runtime, which made it a lot easier to write and maintain a good, robust program. Some call this "B&D programming"; others, myself included, call it common sense. I don't expect to crash my car, but I still wear a seatbelt. I use my turn signal even when I don't think there is a car in the next lane. I'm human. I make mistakes. I try to make sure the damage from my mistakes is limited.
Pascal also encouraged good programming practices in an era where there was still debate over whether good programming practices really mattered. It popularized the idea of teaching structured programming from the start (as opposed to in a footnote on page 378 of the textbook).
While the original Pascal specification made it of limited use for "real world" stuff, adaptations (like Borland's venerable Turbo Pascal) gave you all the power of C or even assembly when you needed it.
These days, most of the lessons that Pascal taught have been learned, and learned well in some new languages. Many of the things learned in the creation and growth of Pascal have also been learned, leading to languages which are all-around better. New ideas (like OO) have taken hold. Better hardware makes things like garbage collection and runtime evaluation a lot more practical.
So the need for Pascal itself, in the present day, is pretty minimal. However, it played a critical role in the evolution of computer programming as a science and as a professional discipline. It was the "first real language" many people learned. And much like a classic car that's been eclipsed by more modern technology, Turbo Pascal still has a certain elegance and appeal to those who knew it. Nostalgia, yes, but good stuff, still.
END. (* PROGRAM *)
I'll try and answer some of your questions, or at least shed some light on them. A lot of this really comes down to "historical reasons" -- it was created that way 30+ years ago, and so we're still stuck with it today. Kinda like the "creat()" function in C/Unix. :-)
:)
.c file for a .h file). It provided a form of encapsulation. If you were distributing a unit, you could distribute just the "interface" part and others could still use the unit.
"...randomly placing components around a window makes it hard to group and line up things..."
I think they expect you to use the alignment tools to fix that up. Like you say, Windoze background. The idea of having software arrange your widgets/controls for you is too foreign.
"...every expression is terminated with a semi-colon, like C, except for the last one in a code-block, which is optiona."
Not quite. In C, semicolons are, indeed, statement terminators. In Pascal, they
are statement separators. That's why you see the behavior you do. For better or worse.
Like you, I took to putting semicolons at the end of most things. I solved the IF problem by using BEGIN/END blocks nearly everywhere. It can be argued that is the right way to go in the long term anyway. Remember, Pascal is designed to encourage good programing practices, and sometimes that increases the short term effort required. Sure, newer languages like Python do a better job, but building Python on the hardware of 30 years ago wouldn't be practical.
"Furthermore, blocks start with 'begin', and end with 'end'. That's alot of characters to type... "
That's why God invented macros.
"Finally, a unit is split up in sections like 'interface', 'implementation'."
Turbo Pascal (the ancestor to Delphi and Object Pascal) created units as a way to easily define libraries. You created an "interface", which was the published API for the library -- kinda like a C header file. The "implementation" was the code (like the
"Why aren't these simply blocks?"
Mainly because they function at a higher level then the normal lexical scopes that BEGIN/END define. In particular, you can define globals that are part of the implementation only, or are also published in the interface.
"And why is the unit itself some sort of half block terminated with 'end.'"
A Pascal program begins with "PROGRAM Foo" and ends with "END."; the Unit syntax just follows suit. No BEGIN was used for the global scope. I expect it's mainly because the "PROGRAM" (or "UNIT") implies you are starting; it also means BEGIN/END are only used to create lexical scopes. The period at the end just signifies the end of the program, same as with an English sentence. It fits Pascal's general approach of trying to provide redundency for safety.
"It's all a matter of taste in the end."
Absolutely.
"They just randomly lock up or reboot at various times. These are office machines, not gaming rigs. The one that is giving me fits right now is a fresh install of XP Pro with updates, not a bit of spyware."
Well, something must be wrong with your setup, because I sure don't have that problem. As much as I hate to say it, when properly administered, on reliable hardware, the Windows XP OS itself is pretty stable.
My job includes the care and feeding of about 50 Windoze workstations. The only time I've seen those kinds of problems has been due to either a hardware fault or bad layered software that was installed. The OS itself is stable.
Of course, this is a strongly managed environment. Users don't have any admin or power rights to their computers; we filter anything that even looks like it might be a virus on before it gets near the users; we filter email and web access; web access is through an authenticated proxy server; we test software before it is installed; etc, etc. We do our homework and then some. We also only buy unexciting computers from trustworthy name-brand vendors; no whiteboxes. Because of that, the OS is stable.
Applications still suck. MS Office seems to find a new way to drive me crazy every day, and our ERP system is clunkier then a cement truck on a race track. But the OS itself is stable and confines the damage to individual applications. For what that's worth.
This ain't cheap. Windows costs way more to do right then other OSes, no doubt about it. But it can be done.
While the OP does sound a bit trollish ("Google who can't figure out... even as good a search as their competitors..."), there are some valid points to be had here.
The reason I (and many others, I'm sure) switched from Alta Vista to Google was that Google consistently returned the results I wanted when given a simple query. AV would generally require me to build a relatively complicated expression (often in several iterations) to get good results. Google just needed a word or two. Google's idea of using links to rate page's value ("PageRank") was a brilliant innovation.
Google is still on track in many ways. You'll notice a lot of their recent features can be described as "figuring out what you want", rather then you *telling* Google want you want. That is, if you enter a street address, it gives you a map. If you enter a person's name and town, it gives you their phone book entry. This is a subtle but very powerful and innovative idea. Most systems make you tell the computer what you want. Even if it's a well-designed UI, you're still picking things from drop-down lists and what have you. Google just has a free-form text-entry field, and figures it out. No syntax. No UI other then the English language. Brilliant.
A lot of Google is also just using the web really well. Look at their recent maps system. Maps have been done before by just about everybody, but nobody had come close to doing it as well as Google. That's not so much innovation as just knowing how to wring every ounce of power out of the popular web browsers and JavaScript, but it still counts as progress.
(Geez, I'm starting to sound like a Microsoft press release with all this "innovation" crap. But I believe it's actually true in this case.)
However, despite all this sycophanthy, I think that Google does have serious room for improvement. For one, the OP is right that search engine spam is starting to seriously diminish the value of Google. That's largely a consequence of success. Just like all the viruses target Windoze because that's where the targets are, SEO's target Google. However, Google still needs to do something about it, or Google will continue to loose value. (Same with Microsoft and the virus problem, incidentally.)
Most of the spam on Google is very obvious. It all looks the same, even in the subject lines. One of the things Google does so well is recognize the same kinds of patterns that humans do. So why can't they introduce some heuristics to filter out sites that are selling things? Make it an option, even, so that people can choose. That will even make a lot of the SEO's happier, as they 'll get people who are interest in buying stuff, and not hits from people who don't want their crap.
The search engine spam problem is doubly-bad with Google because of the fact that their UI is so simple. With Altavista, it might take a more complex query, but at least I had the option. With Google, I often lack the tools (query syntax) needed to refine my search and filter out the spam. Lately I've found myself wishing for the power of Altavista's syntax combined with the human-language intelligence of Google.
I don't want Google to become the Novell of the search engine world. Novell, you may recall, had an innovative, unique, and well-done product (NetWare) when they first started to succeed back in the early 1990s. And the early 1990s is right where Novell and NetWare stayed for at least a decade, while everyone else caught up and then passed them.
It would be a shame for the same thing to happen to Google.