I have a "Download" directory for that:-) Since the browser remembers most of the time the last directory that it was used in, I can save everything in that directory with the added benefit of being able to organise more clearly (Download\Games, Download\Linux, Download\Chat, Download\Tools, Download\pr0n\chick1, etc, etc...) The only reason I use files on the desktop for are shortcuts and small text documents that contain a "ToDo" or "Urls to visit".
Note that a lot of people complain about "disks not being in one root structure" on Windows machines. Now I use this in my advantage. C:\ is for the OS, D:\ is for Data (user files), G:\ is for Games, P:\ is for Programs (by manipulating the registry you can move that stupid C:\Program Files and Common Files) and T:\ is for temporary stuff (including my downloads, so you guessed it it is on T:\Download). Essentially I use disks as super-folders. The added benefit is: if for whatever reason any of these disks (partitions) gets corrupted you still have everything else left. Really, if you lose the OS (C:\), reinstall it you can practically use 85% of all programs without needing a real reinstallation of the application. Works especially well for games and small but usefull tools.
Don't misunderstand me, I *love* the "mount-where-you-need-it" philosophy of unices, but because of my DOS background I still tend to think in partitions and disks. So most of the time, I use/mnt/win/mnt/data/mnt/temp, etc to mount disks. Perhaps the day will come, I'll use a separate disk/partition for my home directory and for my temp directory (that is the two that jump to my mind) and just mount them, but I'm not yet ready for it.
Well, it died a fast but painfull death. One of the reasons I can think of, is that -back in its days- people that used computers came from the DOS world and Windows 3.xx was just getting popular. In both Windows 3.xx and DOS, people were *used* to handle directory structures. They knew how it worked. Honestly, even todays users can handle that kind of complexity. I never ever met a user that wasn't able to organise his stuff in folders (I still call them directories), provided he knew how to make a new folder.
Desktops on the other hand become quickly unusable because of the vast number of icons that people tend to drop on it (shortcuts or not, nothing is worse than seeing a.exe on a desktop) My own strategy is: no more than 7 icons at any given moment on my desktop. It just works best for me.
Note that under Linux I use WindowMaker, which is really an excellent concept (but still very close to the desktop paradigma). The only thing I sometimes miss is that I can't drop a temporary file on the desktop. (under windows I sometimes drop a TXT or URL file to recall me what to do the next time I turn on the computer). Perhaps it is possible, but I don't know how:-)
As for a final note: Microsoft Bob might have had a bright future, but it was 10 years too early. Back then people didn't need the handholding. Now, people who never needed a computer (and are very computer-illiterate) want to use a computer and are confronted with an interface that bears the "history of computing within itself".
I didn't say it was a pipe-dream, you shoudn't twist my words that way. I do realize that more computing power results in faster cracking of ciphers. The discussion *then* is about the computing power of the machine, not about the NP character of the problem. Given an infinitely fast computer with infinite memory any NP problem will be solved in no-time. Since neither is possible, we have to work with approximations of "infinity" which are of course finite, so automatically the calculating time will be longer. So assume you can break a 1024bit key in 1 second because you have more computing power, not because you have found a "better way to solve". No problem, I'll use a 1024^1024 key instead, and you will have a big problem again because your super-fast computer has become slow again. It's that simple. The point is, if you find a way to solve an NP problem *always* in 1 second (or a "short" time), then *I* to have a problem because my cipher has become useless.
The thing that the grand-parent poster did ignore, is the fact that speed of computation linked to a certain machine has nothing to do with complexity of algorithms. Since we are talking complexity in this thread, the remark looked quite clueless. I'm sorry if you don't feel that way.
One point I might add, is that the massively parallel methods like DNA and Quantum computing do circumvent the problem. Assume that checking a certain key takes only one second (heck, it would be fast because it's called "Decoding") and you try all possible keys *at once* (which DNA and Quantum computer do), it would take just 1 second to decipher it. My point is: this is the non-deterministic part of NP, if you go massively parallel, you have no problem with NP.
I don't know if you realise what you are saying: NP-completeness has nothing to do with the computing power of a calculating method. It is intrinsic to an algorithm. The calculations you describe (well, I only know about DNA and quantum computing) are massively parrallel and non-deterministic. That is what the N in NP stands for "Nondeterministic Polynomial (time)", this means, using a non-deterministic calculation method it alread is Polynomial time. So what you describe effectively avoids the NP problem.
Look, it's been a while I had these classes so I might be wrong on some points...but as far as I remember this is right. Feel free to correct me.
The only reason is brands: cars are status symbols and that is the reason people tend to prefer Audi over VW. I do pick those two brands on purpose (I know you used Mercedes), because they use the same platforms. You buy an Audi A3 or Audi TT, but you have exactly the same chassis as a VW Golf. Not many people know this (well, "many" is relative here) Look up the price difference between the cars. I have to admit the Audi finition looks sleeker (a bit), but everything you get in an Audi, you can get it in a VW often even standard or cheaper.
Point is: people look at your car and rate you according to it, they don't look at the software you run on your computer as long as it "works". Actually software is even worse: people actually looked at my Linux desktop and the most common exlamation was "Say, that isn't windows, how can you do any work on it?" Mostly I just smile and don't bother, they don't want to listen anyway.
It's not exactly true that you don't get support, you just have to *do* something yourself for it. That's the way OSS works. Of course, picking up the phone and calling someone is easier. Honestly: if your troubles aren't too complex or too technical in nature, your chances are good that you'll find -after some googeling- a webpage explaining a fix or a workaround. Commercial product support often only give workaround (ever called Oracle?)
Once even I was stuck on an obscure PCMCIA SCSI card, and I complained about it on slashdot. The next day there was a friendly mail from a fellow slashdot reader that explained me to download the pcmcia-cs package instead of relying on the kernel itself for PCMCIA support. I didn't expect this to happen and for this thanks to the slashdot user that helped me (hope you read this!) And this is actually quite a technical question.
I think the technical support that normal user needs, has rarely that kind of complexity. I work at a consulting company and if the secretary (or management/marketing type) has a problem they ask the guys in standby. I guess, in more traditional companies, they call the internal helpdesk and/or ask the IT guy when he emerges from the basement to fetch some coffee. The home user is another type of animal: I never met anyone who actually called the support of Microsoft for help with installing their printer (for example). I reinstall/troubleshoot people (payment: one case of beer and free beer while I'm helping out) and I have seen home configurations with tons of non-working devices on machines ranging from Windows 3.0 to Windows 2000.
For one point: Linux is not yet ready for the desktop in the classic sense (I use it as desktop, but that's not important), but in companies that have well-defined installations and platforms it might stand a chance. Home user desktop: not yet, for that it is too complex to manage... I need to say that it gets better, but it's not there yet.
...but in some regions in France, the imported Bullfrog is slowly starting yet another ecological disaster. Read more about it here . So, yes, the frogs are having a frog problem...sorry, bad pun:-)
You have a point there, nevertheless: I percieved OS/2 as a powerfull alternative to the DOS/WfW combo of the time. It was IBM, but IBM was already crumbling, and, being under the influence of my pa who was reponsible for IT at a bank in that time. Of course IBM was king there. So yes, we had a PS/2 Model 50 back then. Whatever you say aboutthe MCA bus, it was damn good for it's time! Had it not been patent-encumbered, we might all be using MCA derivates nowadays. Hell we even had a Soundblaster MCA, which did very very fine....of course it cost an arm an a limb.
This is sooo sad that it is funny. I was wondering if they actually did link to outside sites *and* if they have written permission themselves? So, after visiting I noticed that no link seem to go outside www.kpmg.com. At least they are consistent.
I hope they do know that many search engines rely on linking to sites: it actually means they want to be an island in the middle of the internet...
Yep, sometimes one needs "extremists" to show a point. You don't get public attention by being tolerant and staying in your corner: Stallman is what I would call a "extremist-opensource-advocate", he went out in public and showed his intolerance for proprietary software.
I would nearly compare it to environmentalists back in the seventies that wanted to banish all industrial activity because of the pollution. (Anyone recall GreenPeace back then?) They made it clear to the world that we were on the wrong track. The world now has taken a moderate standpoint to environmentalism. I think that Stallman has done the same for opensource: the IT world now has embraced it but won't take it to the extreme.
History repeats itself in some form or another. Note for potential flamers: I use the word "extremist" in the context of "someone with extreme viewpoints" not in the current context of "terrorist". Also I wish to note that I have nothing against environemntalists, they need to be there and I respect nature as much as I can.
Great quote. I should remember it! The problem is that most of the time the software that has to be created is a re-hash of an already existing package. I talk about experience here:-)
But I can also agree, some software tends to have "quantum effects"... (I'm going to read the PDF when at home, you know firewalls 'n stuff)
I love Opera too, best browser IMHO that is even useable on very old machines (my Linux machine is a P120 laptop..yes I have better, but I like that one). I actually really like the interface of Opera: it is very userfriendly for my feeling. Yesterday I downloaded 5.05-tp so that I could even use Java support (with the blackdown JRE) but of course that crawls. Mozilla is of course out of the question for that machine, but now I wonder: how does Galeon peform on older machines? Does it still need Mozilla to be installed, or does it come "all-in"?
Ah, just a browser, where has this philosophy gone? When will developers see that all that integration is not needed at all: just like you, I have my own mailclient, (Sylpheed) and for IRC, well, just plain unix-console IRC... What more does one need, eh?
All the points you attribute to "Software Engineers" were covered in detail by my Computer Science study. (Except for the testing, we did some algorithm correctness proving but thats a major PIA)
documentation: Never start coding without specs. This was bashed in us, bashed I say you. Actually, professionally, I know real world doens't work that ideally: sometime you just have to start if 75% of the spec is there because the customer can't decide on the "right" functionality.
Testing: Mostly module testing. Planned: first by the developer, then by a testing team. If very critical by a second testing team. Test cases must be carfully specified and generated from/cross-checked with the specification. I have to admit that most of my testing knowledge comes from my work experience.
Planning: Yes, any SE would do that...so would any CS that is worth is diploma. Data structures are heavily covered in a CS education. (I still have nighmares of 2-3 trees J/K)
Resource use: I hope you do realise that complexity of Algorithms (in Time and Space) is one of the major topics in a CS education.
Yet, I am a computer scientist and proud to be one. Software engineering is part of any CS curriculum. Often people are confused by the term "Computer Scientist" because they do not know the background. Mostly they think I ask me if I'm some kind of engineer. I tell them I am not, and that I'm "just" a computer scientist. Well, most of the time they look as if I said "mad scientist", but that is another story.
Interesting, I actually do not know any firefighters, policemen or military people. So your point may stand. My brother however is a money-transporter (I have no clue how that is called in english) which is close to a paramilitary organisation. As far as I can see, there isn't much of socializing getting on. Oh, they are stong together when one of them gets shot again (happened trice in the last year), but apart from that...I don't think so.
My dad has been a normal employee ranging from accountant to auditor and I don't think he went often to parties or that we invited anyone anytime. Hey, perhaps it's just a thing in upbringing?
Note that I think that a lot of people I met in IT tend to be very very individualistic which would probably interfere with what one could call "typical social bonding".
Exactly! The day I started working, I said: no mixing up of worklife and private life. But then I'm considered as a lonely shark and have no social life:-)
To be on topic again, I even think that those stereotypical bonding-rituals described in the "Ask Slashdot", don't really exist. It's all TV and you shouldn't believe what is on TV.;-)
Nice that TweakUI does that now for the unwashed masses, but I did that since the late '95:-) Good ole regedit (or regedt32 for those that want to manage rights under NT) did the job fine for me. It takes a while after initial install (seaching/replacing keys) but I think it is worth it.
I do similar stuff with C:\Program Files because on my machines it is E:\WinApp. You won't find anything like "Program Files" on my machines.
You have a point that "folder" is a more simple word, but consider this: you do helpdesk work and support a user, now you have to say "click on the My Documents folder", that's much longer than "click on the docs directory":-) Of course "click on the docs folder" is even more short. Honestly, it *still* is a directory: under the shell the command is still "dir" and the directories as still marked "<DIR>", same under Linux (or any other Unix), "ls -la" will give you "drwxrwxrwx", the first one is a 'd' for directoy not an 'f' for folder. We should keep the traditional jargon, and not switch it every 2 years because "users don't grasp it". Let's just agree that "folder" is the GUI representation of directory and everybody is happy;-)
Yes, I know new Windows versions do that: this goes up to the point that users think that you can only store images in the "My Pictures" folder and only music in the "My Music". Who was the "genius" who came up with the idea of putting "My" in front of every directory name, anyway? It makes me wanna puke.
Creating such annoying directories (which you cannot delete, since windows insists on recreating them) just puts a small and ineffective patch on the large wound which is called "computer illiteracy". That should be fixed, not the computers.
So I agree with you on one point: organize your files like you want. I don't like the default Windows "My Document" folder and by hacking the registry it is very well possible to use for example D:\Users\%USERNAME% instead, which corresponds better to my way of thinking. Under Linux I just juse/home/$USER because I like that structure.
The problem how to manage your files is beyond that for "joe sixpack". A lot of end-users don't even grasp the concept of a directory (or as it is now called "folder", what happened to the old terminology?). Honestly, I have seen users that save everything in their "My Documents" folder, and that includes their 1200 MP3's, their 50 Word documents, their 20 Excell documents and the 500 jpegs of pr0n. All flat in one directory, I kid you not! If you even try to explain what a "Folder" is, they freak out that you are being to technical. If by accident an application does not save their stuff in "My Documents", well they are hosed because they will never find it back. Sad, isn't it?
Well, I upgraded yesterday to 2000b (not 2001b, I was on #ICQHelp and they said it was utter crap) and it has the same issues and the ads don't make it prettier. Besides: protocol's shouldn't change! I don't complain I can't do SMS and all the new fancy stuff: I'm more than happy with basic messenging....stuff that was supported since pre-98a! Honestly: I do not upgrade software when it "Works For Me". The beta-argument is complete idiocy: they haven't change it in 2 years and now suddenly it is obsoleted? Sure, and I am Madonna!
Do you realise how heavy the new crapware is they release? Do I have to buy a new computer just to be able to do what I did more than 2 years ago with perfectly working software. I cannot buy a top notch PC for everyone in my family, so my sis has to do with a P166 and guess who uses ICQ most? Yes, my sis....
How can reverting to TCP-only (it probably is TCP only, because UDP is connectionless and doesn't ensure that packets arrive at destination) when the older version support BOTH? If the older versions support both they should be able to communicate with the newer ones reverting to TCP... Besides, there has always been "server over server" and now even that doesn't work. Sorry, I don't grasp your logic: if it wasn't for your low ID I'd suspect you for posting this as pure and utter flamebait.
Considering how *unreliable* ICQ has become, I frankly doubt that this would have any use. Honestly since some time it has become impossible to communicate with people running newer versions of ICQ (still using 99b-Rev A here, or LICQ).
Besides, I know it's possible to do ICQ on handhelds for a long time. I have a Psion and there is an ICQ client available. It is paying so I never bothered. (Use google to find it) I've used Opera on my Psion for the sake of it and that works great, if this is some kind of integrated Phone/Psion I could get interesed (including speadsheet, Contacts, Word, Jotter,...) I always have looked down on Palm owners, because the Psion in it's many incarnations is really superior IMHO. Too bad Psion stopped making hardware. As for Nokia hardware, I alway found them "feeling" cheap, more like toys...Give me a good Siemens anyday.
Slashdot Mirror
The only reason I use files on the desktop for are shortcuts and small text documents that contain a "ToDo" or "Urls to visit".
Note that a lot of people complain about "disks not being in one root structure" on Windows machines. Now I use this in my advantage. C:\ is for the OS, D:\ is for Data (user files), G:\ is for Games, P:\ is for Programs (by manipulating the registry you can move that stupid C:\Program Files and Common Files) and T:\ is for temporary stuff (including my downloads, so you guessed it it is on T:\Download). Essentially I use disks as super-folders. The added benefit is: if for whatever reason any of these disks (partitions) gets corrupted you still have everything else left. Really, if you lose the OS (C:\), reinstall it you can practically use 85% of all programs without needing a real reinstallation of the application. Works especially well for games and small but usefull tools.
Don't misunderstand me, I *love* the "mount-where-you-need-it" philosophy of unices, but because of my DOS background I still tend to think in partitions and disks. So most of the time, I use /mnt/win /mnt/data /mnt/temp, etc to mount disks. Perhaps the day will come, I'll use a separate disk/partition for my home directory and for my temp directory (that is the two that jump to my mind) and just mount them, but I'm not yet ready for it.
Desktops on the other hand become quickly unusable because of the vast number of icons that people tend to drop on it (shortcuts or not, nothing is worse than seeing a .exe on a desktop) My own strategy is: no more than 7 icons at any given moment on my desktop. It just works best for me.
Note that under Linux I use WindowMaker, which is really an excellent concept (but still very close to the desktop paradigma). The only thing I sometimes miss is that I can't drop a temporary file on the desktop. (under windows I sometimes drop a TXT or URL file to recall me what to do the next time I turn on the computer). Perhaps it is possible, but I don't know how :-)
As for a final note: Microsoft Bob might have had a bright future, but it was 10 years too early. Back then people didn't need the handholding. Now, people who never needed a computer (and are very computer-illiterate) want to use a computer and are confronted with an interface that bears the "history of computing within itself".
Given an infinitely fast computer with infinite memory any NP problem will be solved in no-time. Since neither is possible, we have to work with approximations of "infinity" which are of course finite, so automatically the calculating time will be longer. So assume you can break a 1024bit key in 1 second because you have more computing power, not because you have found a "better way to solve". No problem, I'll use a 1024^1024 key instead, and you will have a big problem again because your super-fast computer has become slow again. It's that simple. The point is, if you find a way to solve an NP problem *always* in 1 second (or a "short" time), then *I* to have a problem because my cipher has become useless.
The thing that the grand-parent poster did ignore, is the fact that speed of computation linked to a certain machine has nothing to do with complexity of algorithms. Since we are talking complexity in this thread, the remark looked quite clueless. I'm sorry if you don't feel that way.
One point I might add, is that the massively parallel methods like DNA and Quantum computing do circumvent the problem. Assume that checking a certain key takes only one second (heck, it would be fast because it's called "Decoding") and you try all possible keys *at once* (which DNA and Quantum computer do), it would take just 1 second to decipher it. My point is: this is the non-deterministic part of NP, if you go massively parallel, you have no problem with NP.
The calculations you describe (well, I only know about DNA and quantum computing) are massively parrallel and non-deterministic. That is what the N in NP stands for "Nondeterministic Polynomial (time)", this means, using a non-deterministic calculation method it alread is Polynomial time. So what you describe effectively avoids the NP problem.
Look, it's been a while I had these classes so I might be wrong on some points...but as far as I remember this is right. Feel free to correct me.
Point is: people look at your car and rate you according to it, they don't look at the software you run on your computer as long as it "works". Actually software is even worse: people actually looked at my Linux desktop and the most common exlamation was "Say, that isn't windows, how can you do any work on it?" Mostly I just smile and don't bother, they don't want to listen anyway.
Honestly: if your troubles aren't too complex or too technical in nature, your chances are good that you'll find -after some googeling- a webpage explaining a fix or a workaround. Commercial product support often only give workaround (ever called Oracle?)
Once even I was stuck on an obscure PCMCIA SCSI card, and I complained about it on slashdot. The next day there was a friendly mail from a fellow slashdot reader that explained me to download the pcmcia-cs package instead of relying on the kernel itself for PCMCIA support. I didn't expect this to happen and for this thanks to the slashdot user that helped me (hope you read this!) And this is actually quite a technical question.
I think the technical support that normal user needs, has rarely that kind of complexity. I work at a consulting company and if the secretary (or management/marketing type) has a problem they ask the guys in standby. I guess, in more traditional companies, they call the internal helpdesk and/or ask the IT guy when he emerges from the basement to fetch some coffee.
The home user is another type of animal: I never met anyone who actually called the support of Microsoft for help with installing their printer (for example). I reinstall/troubleshoot people (payment: one case of beer and free beer while I'm helping out) and I have seen home configurations with tons of non-working devices on machines ranging from Windows 3.0 to Windows 2000.
For one point: Linux is not yet ready for the desktop in the classic sense (I use it as desktop, but that's not important), but in companies that have well-defined installations and platforms it might stand a chance. Home user desktop: not yet, for that it is too complex to manage... I need to say that it gets better, but it's not there yet.
...but in some regions in France, the imported Bullfrog is slowly starting yet another ecological disaster. Read more about it here . So, yes, the frogs are having a frog problem...sorry, bad pun :-)
You have a point there, nevertheless: I percieved OS/2 as a powerfull alternative to the DOS/WfW combo of the time. It was IBM, but IBM was already crumbling, and, being under the influence of my pa who was reponsible for IT at a bank in that time. Of course IBM was king there. So yes, we had a PS/2 Model 50 back then. Whatever you say aboutthe MCA bus, it was damn good for it's time! Had it not been patent-encumbered, we might all be using MCA derivates nowadays. Hell we even had a Soundblaster MCA, which did very very fine....of course it cost an arm an a limb.
I hope they do know that many search engines rely on linking to sites: it actually means they want to be an island in the middle of the internet...
I would nearly compare it to environmentalists back in the seventies that wanted to banish all industrial activity because of the pollution. (Anyone recall GreenPeace back then?) They made it clear to the world that we were on the wrong track. The world now has taken a moderate standpoint to environmentalism.
I think that Stallman has done the same for opensource: the IT world now has embraced it but won't take it to the extreme.
History repeats itself in some form or another.
Note for potential flamers: I use the word "extremist" in the context of "someone with extreme viewpoints" not in the current context of "terrorist". Also I wish to note that I have nothing against environemntalists, they need to be there and I respect nature as much as I can.
But I can also agree, some software tends to have "quantum effects"... (I'm going to read the PDF when at home, you know firewalls 'n stuff)
Mozilla is of course out of the question for that machine, but now I wonder: how does Galeon peform on older machines? Does it still need Mozilla to be installed, or does it come "all-in"?
Ah, just a browser, where has this philosophy gone? When will developers see that all that integration is not needed at all: just like you, I have my own mailclient, (Sylpheed) and for IRC, well, just plain unix-console IRC... What more does one need, eh?
Yet, I am a computer scientist and proud to be one. Software engineering is part of any CS curriculum. Often people are confused by the term "Computer Scientist" because they do not know the background. Mostly they think I ask me if I'm some kind of engineer. I tell them I am not, and that I'm "just" a computer scientist. Well, most of the time they look as if I said "mad scientist", but that is another story.
"It's better to have loved and lost than never to have loved"
The latter is my case, so I bet I don't know what I'm talking about. It's a cliché anyway :-/
You mean those come in combo packs? ;-)
Since that is not going to happen, I'd really like a nice and shiny iBook (~$1500).
I'm on my way.
My dad has been a normal employee ranging from accountant to auditor and I don't think he went often to parties or that we invited anyone anytime. Hey, perhaps it's just a thing in upbringing?
Note that I think that a lot of people I met in IT tend to be very very individualistic which would probably interfere with what one could call "typical social bonding".
To be on topic again, I even think that those stereotypical bonding-rituals described in the "Ask Slashdot", don't really exist. It's all TV and you shouldn't believe what is on TV. ;-)
I do similar stuff with C:\Program Files because on my machines it is E:\WinApp. You won't find anything like "Program Files" on my machines.
You have a point that "folder" is a more simple word, but consider this: you do helpdesk work and support a user, now you have to say "click on the My Documents folder", that's much longer than "click on the docs directory" :-) Of course "click on the docs folder" is even more short. ;-)
Honestly, it *still* is a directory: under the shell the command is still "dir" and the directories as still marked "<DIR>", same under Linux (or any other Unix), "ls -la" will give you "drwxrwxrwx", the first one is a 'd' for directoy not an 'f' for folder. We should keep the traditional jargon, and not switch it every 2 years because "users don't grasp it". Let's just agree that "folder" is the GUI representation of directory and everybody is happy
Creating such annoying directories (which you cannot delete, since windows insists on recreating them) just puts a small and ineffective patch on the large wound which is called "computer illiteracy". That should be fixed, not the computers.
(At least I'm not alone with my feelings).
So I agree with you on one point: organize your files like you want. I don't like the default Windows "My Document" folder and by hacking the registry it is very well possible to use for example D:\Users\%USERNAME% instead, which corresponds better to my way of thinking. Under Linux I just juse /home/$USER because I like that structure.
The problem how to manage your files is beyond that for "joe sixpack". A lot of end-users don't even grasp the concept of a directory (or as it is now called "folder", what happened to the old terminology?). Honestly, I have seen users that save everything in their "My Documents" folder, and that includes their 1200 MP3's, their 50 Word documents, their 20 Excell documents and the 500 jpegs of pr0n. All flat in one directory, I kid you not! If you even try to explain what a "Folder" is, they freak out that you are being to technical. If by accident an application does not save their stuff in "My Documents", well they are hosed because they will never find it back. Sad, isn't it?
s/"server over server"/"send over server"/g
Do you realise how heavy the new crapware is they release? Do I have to buy a new computer just to be able to do what I did more than 2 years ago with perfectly working software. I cannot buy a top notch PC for everyone in my family, so my sis has to do with a P166 and guess who uses ICQ most? Yes, my sis....
How can reverting to TCP-only (it probably is TCP only, because UDP is connectionless and doesn't ensure that packets arrive at destination) when the older version support BOTH? If the older versions support both they should be able to communicate with the newer ones reverting to TCP... Besides, there has always been "server over server" and now even that doesn't work. Sorry, I don't grasp your logic: if it wasn't for your low ID I'd suspect you for posting this as pure and utter flamebait.
Considering how *unreliable* ICQ has become, I frankly doubt that this would have any use. Honestly since some time it has become impossible to communicate with people running newer versions of ICQ (still using 99b-Rev A here, or LICQ).
Besides, I know it's possible to do ICQ on handhelds for a long time. I have a Psion and there is an ICQ client available. It is paying so I never bothered. (Use google to find it) I've used Opera on my Psion for the sake of it and that works great, if this is some kind of integrated Phone/Psion I could get interesed (including speadsheet, Contacts, Word, Jotter,...) I always have looked down on Palm owners, because the Psion in it's many incarnations is really superior IMHO. Too bad Psion stopped making hardware.
As for Nokia hardware, I alway found them "feeling" cheap, more like toys...Give me a good Siemens anyday.