The capacitors have not dried out. Since they use a water based electrolyte, that could be the most critical point. Sometimes they dry out after just 2 years of normal use, due to the higher temperature during operation.
But the most important question would be: why? Do you really think that in 50 years anybody cares about a PC that was mediocre in the year 2000? Very few people get excited about punch cards, and that will be exactly how CDs will feel to someone used to wireless solid state data storage.
> Why not go after the big offenders for power usage? Lighting and heating? Or the stupidly wasteful standby modes on many TVs, computers and appliances?
They do. Standby is handled in the same piece of legislation, which an eventual limit of 0.5W. Lighting also falls under the same rules, with the least energy efficient product (the 25W opaque incandescent light bulb and its friends) bound to disappear from the market.
Heating is dealt with in national legislation, and for example in Germany condensing boilers are pretty much mandated, as is improved thermal insulation.
> The fact that only the display and printer are mapped back to the client (and we use the upd, no native drivers) means there's not really any exposure to client malware.
Yes, but what about user input? Malware could easily intercept key strokes, and that could be sensitive information. Do you use passwords, for example? I know single sign on is the big thing, but I have not seen a single place where it actually works.
> If my desktop doesn't work, how much is the company spending for me to sit around doing nothing.
Exactly. Ask your "customers" how long it would take them to do the same job. That would usually be a lot longer than it takes you - and bingo, you have a business case.
> Sometimes it's important to know the simple solution as well. Most (all?) of the algorithms that perform better than O(n^2) have a fairly high constant attached to them
Very true. If you get (nearly) nothing across, but they understand KISS at the end of the course, it would still be a success.
Second thought: programming is all about structure. How does the problem feel, and "why is this solution so awkward?" Try to at introduce different programming paradigms, at least give them an idea of it. Every programmer needs to know about functional programming, procedural programming, and nowadays also object oriented.
And if they understand the different between interpreted and compiled, data and program, then I think you have put down a pretty solid foundation.
> Or you could look at something like Zimbra, which has all your mail basics plus extra goodness like calendaring built-in.
I think calendaring is an extremely important part of the solution. University life revolves around lectures, meetings and deadlines, and being able to organise these properly can save a lot of time and confusion. So how first question to ask is: how would students interact with and benefit from the calendering system?
The next big question is how it links in with other systems. A lot of systems will interact with the calendaring. Attachments relate to your file servers, at least a lot of data will go back and forth. Can you tie all this together? As the OP said, it is easier with a system that you run yourself, but there is generally no reason you cannot do it with a "web service".
> I'd suggest a brief talk on satellites and then show them Google Earth.
I am not sure the satellites are essential, but Google Earth is a great idea. It shows how "software as a service" works, and why servers are useful. If the OP gets the concepts of hardware/software and client/server across, that should be a good result for 20 minutes.
> For decades all developed countries have kept visa information on centralized databases, thats the whole point of issuing visas!
Yes, you would think so, but actually the UK did not. Even visas issued last year are not in the current database. And the UK does no exit checks, which means they have no idea how many people on visas are still in the country, or have since left.
So yes, the whole exercise is about building a big database. Why this one should work any better than previous attempts is beyond me, though.
True, but why don't they use two read heads, one on each side of the drive? That seems a lot easier than going into rocket science material with 25000rpm. After all, at that speed centrifugal forces are over 5 times of the value at 10000 rpm.
> I don't know if this is the case for you, but I find most people who find LaTeX hard are using it wrongly.
Very true words. The whole idea of LaTeX is that it does all the formatting work for you. So complaining about "complexity" is really missing the issue: LaTeX is as complex as necessary for the task. Use it wisely, and it will go a long way.
If you do not want to deal with the complexity of different styles, then a front end like LyX can hide a lot of it. You still get high quality results, you can switch between styles, and you can use additional features manually if necessary. This does not mean that LyX is without fault, but I think it is a step in the right direction (very much unlike Word).
Concerning the OP's question about a document processor without the "compiler metaphor" (and it is a paradigm, not a metaphor)... there is no such thing. The whole idea of a document processor is that things are done right, and not fast. Doing this in real time is just asking for trouble. So you either end up with a draft view as in LyX, or with a sluggish real time preview (as you find in a few LaTeX editors). Anyway, with a document processor you are supposed to put down semantics, and not form, so looking at the exact final form is wasted precision. If you want to have certain things in certain places, LaTeX has commands and overrides to achieve that.
> IMO, the equation editor in Word 2007 was a huge improvement over the previous versions.
Indeed, but it is still not as fast to use as LyX, and the results are not as good. So while I could imagine using it if necessary (which I never considered with previous version), it is still not my preference.
> I'm trying to run Ubunu on a VIA epia for some time now, but their graphics solution is as unstable as hell.
Yep, same here. The rest of the board works quite well, but I had endless trouble with the built in graphics. With Ubuntu 8.04 it seems to work ok now, but the picture is still a bit fuzzy.
> but what happens when you need those same emails that are over 180 days old that would have EXONERATED you?
Well, obviously this company has decided that old emails are much more likely to work against them, and this even overrides the loss of productivity due to important emails going missing etc. I really wonder what kind of business this company is in, and what their business strategy is:-(
Or maybe it is just one CEO that knows something funny went on, and now he/she is trying to destroy the evidence whatever the cost.
> Of course, the real problem is the idea that users manually editing a 32MB file on a regular basis is at all a sensible idea.
The problem is a combination of two unusual requirements: big files and on-line change tracking.
Anyway, I think the big file size is the more interesting issue. If you have a file with 32 MB size, it should have a very clearly defined structure - it is data. So the changes should also be presented in this structure, which means you need a custom tool. If your file does not have a very clear structure, you have lost already, and all the reviews in the world are not going to help.
> he should have provided a means to allow some trusted individual the means to access these systems
Exactly, this is all about trust. As a system administrator, you have to trust your technology and your colleagues/supervisors. Without a good amount of trust, you cannot be the link that you need to be. A paranoid person cannot do this job.
Of course this also depends a lot on the supervisor, and that seems to be a big of the problem.
That, and rather unclear requirements. So this is a great starting position for a never-ending fiddling project. I hope that is what the OP is after.
Anyway, I would look into some kind of suspend mode. Even old PCs usually have one, although it may or may not work. The other option is to use the suspend function in VMWare, but you still need to get a system up to run VMWare, so it is going to take a few seconds longer.
Of course there are also cheap embedded PC boards available, which would probably solve a whole lot of problems. But if the OP does not want to use them, that is fine by me:-).
> The name "gmail" was already taken for an e-mail provider.
If only it were. G-Mailer is a snail-mailer with an electronic interface, that also handles e-mail, marketed for businesses. So while it can be used to handle email, there is very little danger of it being confused with GoogleMail. I for one am still baffled by the court verdict, but I have to concede that it is "legal".
> Modern life has reached an annoying level of materialism....
> And yet... I have stuff that I really like. I've devoted a lot of time to crafting a PVR from scratch, and having it loaded with thousands of mp3's and hooked up to a nice stereo is something I really enjoy.
No need to blame it on materialism. Meaning is often carrier, transported or represented by material objects. The bible is one example (the physical book), your PVR is another or a marriage picture. This is not necessarily a sign of extreme materialism, it is just the way things work.
As long as meaning cannot stand on its own ("purely spiritual beings"), it must have some manifestation in the real world. Nothing wrong with that, unless you have to move:-)
> In fact I have trouble thinking of an example that fits your claim. Basically the key to British (and most) fame is to be famous in your lifetime first.
What about Alan Turing? Of course he is still much better received abroad than in his own country, but he is a perfect example of an unrecognised genius. He was used to win the war, and then dumped like a hot potato.
I am sad, too, because this was a really useful utility. Something similar would be very much appreciated, and it is not just about bookmarks, but also because of the cookies. Now to be honest, I do not need them synchronised every time, but every once in a while would be nice.
I guess one for the problems with Google Browser Sync is that it had quite a few issues. I lost my cookies several times (in combination with my laptop running out of power). Also it did not work well with some other extensions (AdBlock Plus, I think, or maybe NoScript), creating gigabytes of log file entries. And finally it did slow down the browser. So it was a good idea, but probably not the correct implementation. If Firefox 4 has something even remotely similar, I think we will all be better off.
Actually, Flash doesn't work very well. It still crashes, and running it in a 64-bit Firefox is a pain. Flash works quite well, certainly compared to *cough* Java applets. A 64bit Java plugin is "being planned". And I thought this was 2008...
> Its only problems, as far as I can see, was that it was initially slow
No, the problem is that it is always slow. The current JVM takes 10 seconds to start, and my PC is only a year old. Heck, JVM 1.1 was faster on the hardware common back then. Plus the JVM is my number on cause of crashes, although I hardly use it. And for my Ubuntu/amd64, I cannot even get a native JVM, because 64bit is "in the making" for a decade now.
Or to put it simple: Java is hated because Sun just does not get it.
Slashdot Mirror
> You're assuming that in 50 years:
The capacitors have not dried out. Since they use a water based electrolyte, that could be the most critical point. Sometimes they dry out after just 2 years of normal use, due to the higher temperature during operation.
But the most important question would be: why? Do you really think that in 50 years anybody cares about a PC that was mediocre in the year 2000? Very few people get excited about punch cards, and that will be exactly how CDs will feel to someone used to wireless solid state data storage.
> Why not go after the big offenders for power usage? Lighting and heating? Or the stupidly wasteful standby modes on many TVs, computers and appliances?
They do. Standby is handled in the same piece of legislation, which an eventual limit of 0.5W. Lighting also falls under the same rules, with the least energy efficient product (the 25W opaque incandescent light bulb and its friends) bound to disappear from the market.
Heating is dealt with in national legislation, and for example in Germany condensing boilers are pretty much mandated, as is improved thermal insulation.
> Making a browser that integrates Java in a reasonable way
No, it is not. My browser starts faster than the JVM - and I would like to keep it that way.
> The fact that only the display and printer are mapped back to the client (and we use the upd, no native drivers) means there's not really any exposure to client malware.
Yes, but what about user input? Malware could easily intercept key strokes, and that could be sensitive information. Do you use passwords, for example? I know single sign on is the big thing, but I have not seen a single place where it actually works.
> If my desktop doesn't work, how much is the company spending for me to sit around doing nothing.
Exactly. Ask your "customers" how long it would take them to do the same job. That would usually be a lot longer than it takes you - and bingo, you have a business case.
And of course on board of Soyuz you are a cosmonaut, not an astronaut. Space may be the same for everyone, but marketing divides it up :-)
> Sometimes it's important to know the simple solution as well. Most (all?) of the algorithms that perform better than O(n^2) have a fairly high constant attached to them
Very true. If you get (nearly) nothing across, but they understand KISS at the end of the course, it would still be a success.
Second thought: programming is all about structure. How does the problem feel, and "why is this solution so awkward?" Try to at introduce different programming paradigms, at least give them an idea of it. Every programmer needs to know about functional programming, procedural programming, and nowadays also object oriented.
And if they understand the different between interpreted and compiled, data and program, then I think you have put down a pretty solid foundation.
> Or you could look at something like Zimbra, which has all your mail basics plus extra goodness like calendaring built-in.
I think calendaring is an extremely important part of the solution. University life revolves around lectures, meetings and deadlines, and being able to organise these properly can save a lot of time and confusion. So how first question to ask is: how would students interact with and benefit from the calendering system?
The next big question is how it links in with other systems. A lot of systems will interact with the calendaring. Attachments relate to your file servers, at least a lot of data will go back and forth. Can you tie all this together? As the OP said, it is easier with a system that you run yourself, but there is generally no reason you cannot do it with a "web service".
> I'd suggest a brief talk on satellites and then show them Google Earth.
I am not sure the satellites are essential, but Google Earth is a great idea. It shows how "software as a service" works, and why servers are useful. If the OP gets the concepts of hardware/software and client/server across, that should be a good result for 20 minutes.
> For decades all developed countries have kept visa information on centralized databases, thats the whole point of issuing visas!
Yes, you would think so, but actually the UK did not. Even visas issued last year are not in the current database. And the UK does no exit checks, which means they have no idea how many people on visas are still in the country, or have since left.
So yes, the whole exercise is about building a big database. Why this one should work any better than previous attempts is beyond me, though.
> Two Words - Rotational Latency
True, but why don't they use two read heads, one on each side of the drive? That seems a lot easier than going into rocket science material with 25000rpm. After all, at that speed centrifugal forces are over 5 times of the value at 10000 rpm.
> I don't know if this is the case for you, but I find most people who find LaTeX hard are using it wrongly.
Very true words. The whole idea of LaTeX is that it does all the formatting work for you. So complaining about "complexity" is really missing the issue: LaTeX is as complex as necessary for the task. Use it wisely, and it will go a long way.
If you do not want to deal with the complexity of different styles, then a front end like LyX can hide a lot of it. You still get high quality results, you can switch between styles, and you can use additional features manually if necessary. This does not mean that LyX is without fault, but I think it is a step in the right direction (very much unlike Word).
Concerning the OP's question about a document processor without the "compiler metaphor" (and it is a paradigm, not a metaphor)... there is no such thing. The whole idea of a document processor is that things are done right, and not fast. Doing this in real time is just asking for trouble. So you either end up with a draft view as in LyX, or with a sluggish real time preview (as you find in a few LaTeX editors). Anyway, with a document processor you are supposed to put down semantics, and not form, so looking at the exact final form is wasted precision. If you want to have certain things in certain places, LaTeX has commands and overrides to achieve that.
> IMO, the equation editor in Word 2007 was a huge improvement over the previous versions.
Indeed, but it is still not as fast to use as LyX, and the results are not as good. So while I could imagine using it if necessary (which I never considered with previous version), it is still not my preference.
> I'm trying to run Ubunu on a VIA epia for some time now, but their graphics solution is as unstable as hell.
Yep, same here. The rest of the board works quite well, but I had endless trouble with the built in graphics. With Ubuntu 8.04 it seems to work ok now, but the picture is still a bit fuzzy.
> but what happens when you need those same emails that are over 180 days old that would have EXONERATED you?
Well, obviously this company has decided that old emails are much more likely to work against them, and this even overrides the loss of productivity due to important emails going missing etc. I really wonder what kind of business this company is in, and what their business strategy is :-(
Or maybe it is just one CEO that knows something funny went on, and now he/she is trying to destroy the evidence whatever the cost.
> Of course, the real problem is the idea that users manually editing a 32MB file on a regular basis is at all a sensible idea.
The problem is a combination of two unusual requirements: big files and on-line change tracking.
Anyway, I think the big file size is the more interesting issue. If you have a file with 32 MB size, it should have a very clearly defined structure - it is data. So the changes should also be presented in this structure, which means you need a custom tool. If your file does not have a very clear structure, you have lost already, and all the reviews in the world are not going to help.
> he should have provided a means to allow some trusted individual the means to access these systems
Exactly, this is all about trust. As a system administrator, you have to trust your technology and your colleagues/supervisors. Without a good amount of trust, you cannot be the link that you need to be. A paranoid person cannot do this job.
Of course this also depends a lot on the supervisor, and that seems to be a big of the problem.
> You have unrealistic expectations.
That, and rather unclear requirements. So this is a great starting position for a never-ending fiddling project. I hope that is what the OP is after.
Anyway, I would look into some kind of suspend mode. Even old PCs usually have one, although it may or may not work. The other option is to use the suspend function in VMWare, but you still need to get a system up to run VMWare, so it is going to take a few seconds longer.
Of course there are also cheap embedded PC boards available, which would probably solve a whole lot of problems. But if the OP does not want to use them, that is fine by me :-).
> The name "gmail" was already taken for an e-mail provider.
If only it were. G-Mailer is a snail-mailer with an electronic interface, that also handles e-mail, marketed for businesses. So while it can be used to handle email, there is very little danger of it being confused with GoogleMail. I for one am still baffled by the court verdict, but I have to concede that it is "legal".
> Modern life has reached an annoying level of materialism. ...
> And yet... I have stuff that I really like. I've devoted a lot of time to crafting a PVR from scratch, and having it loaded with thousands of mp3's and hooked up to a nice stereo is something I really enjoy.
No need to blame it on materialism. Meaning is often carrier, transported or represented by material objects. The bible is one example (the physical book), your PVR is another or a marriage picture. This is not necessarily a sign of extreme materialism, it is just the way things work.
As long as meaning cannot stand on its own ("purely spiritual beings"), it must have some manifestation in the real world. Nothing wrong with that, unless you have to move :-)
> In fact I have trouble thinking of an example that fits your claim. Basically the key to British (and most) fame is to be famous in your lifetime first.
What about Alan Turing? Of course he is still much better received abroad than in his own country, but he is a perfect example of an unrecognised genius. He was used to win the war, and then dumped like a hot potato.
I am sad, too, because this was a really useful utility. Something similar would be very much appreciated, and it is not just about bookmarks, but also because of the cookies. Now to be honest, I do not need them synchronised every time, but every once in a while would be nice.
I guess one for the problems with Google Browser Sync is that it had quite a few issues. I lost my cookies several times (in combination with my laptop running out of power). Also it did not work well with some other extensions (AdBlock Plus, I think, or maybe NoScript), creating gigabytes of log file entries. And finally it did slow down the browser. So it was a good idea, but probably not the correct implementation. If Firefox 4 has something even remotely similar, I think we will all be better off.
> Its only problems, as far as I can see, was that it was initially slow
No, the problem is that it is always slow. The current JVM takes 10 seconds to start, and my PC is only a year old. Heck, JVM 1.1 was faster on the hardware common back then. Plus the JVM is my number on cause of crashes, although I hardly use it. And for my Ubuntu/amd64, I cannot even get a native JVM, because 64bit is "in the making" for a decade now.
Or to put it simple: Java is hated because Sun just does not get it.
> I feel safer killing insurgents in their backyard rather than killing them here, but I am probably strange that way.
And yet you are here? Why aren't you killing insurgents in their backyard? Maybe because it does not feel all that safe after all...