Frankly, I am tired of endless learning of OS quirks, Windows or some Linux distrubution. I would really like to see computers turn into appliances, with minimal differences between OS's and app's. I help keep up a mixed bag of computers running from P2's to Gateway E-4500's. My eyes cross when going from W98 to W2000 to XP (ME has died the real death). Microsoft certainly is anything but consistent.

So I take it you will buy into the Google ideology that computing in the future will be run from the web browser over internetworks, and hence will be independant of the OS?

Dave

 

Hi Dave,

That assumes one has access to the internet at greater than 56K modem speeds (32.2 Kbaud for me).

Will Google, or any other profit-making entity protect intellectual property when someone utilizes their office suite apps? Believe that, and I want to interest you in some Mississippi Gulf Coast property.

Any modern computer has more power than the old Cray supercomputers, and more storage than anyone might sensibily make use of. It should be an appliance. Anyone should be able to get apps that will do what they need done. It should be seen as a waste of time and energy to have to master multiple pissy OS's to get one's computational needs satisfied.

Of course, this assumes that some agreement of just what constitutes an operating system will ever be arrived at.

 

Hi Bill,

But isn't that the Google ideology, that the OS will become superfluous. All applications will be run from the web browser, and the underlying architecture of the applications on the remote server will be completely transparent to the end user (at least more so than is the case today). There will be no need to master the OS, just the application.

I'm not saying I personally buy into this ideology, but it has been floated around as a potential way forward in the future.

You are right to make the comment regarding the need for Internet access at rates greater than 56k, although many would argue that broadband is increasingly becoming the norm and hence would be able to support this idea of network run PCs. This is obviously not the case for those in remote areas.

As for the mastering of multiple OSes, well some would argue this is the result of divergence from standards and the fact we live with a PC monopoly. In a perfect world all things would work on all computers and the only difference would down to the end users choice, but this is big business and the ideas of embrace and extend means that the available options are often very different.

Perhaps we should all just outgrow computers and get back to entertaining oursleves with a rubix cube!!

Dave

 

Wasn't this what Oracle was pushing a decade or so back? (OS on a web server - not the Rubik's cube)

 

Hi Lee,

Thats an interesting point, however the idea of an OS on a web server and running applications through the web browser is something I have only heard about in recent years. I would interested in seeing anything that suggests this idea may go back many more years, or in any way is related to Oracle.

Dave

 

Network Computer (NC), initiated by Larry Ellison CEO of Oracle, is a similar concept, but I don't think the applications are run on web browser like Google's vision. It's more like the old mainframe idea, where the NCs act as terminals and also centralised applications repository that can be rented by users, technically similar to Java app. It was back in 95 I think, and a few months ago the idea was doing the round again.

Google for Network Computer and Larry Ellison, you would get many many hits.

 

Network Computer (NC), initiated by Larry Ellison CEO of Oracle, is a similar concept, but I don't think the applications are run on web browser like Google's vision. It's more like the old mainframe idea, where the NCs act as terminals and also centralised applications repository that can be rented by users, technically similar to Java app. It was back in 95 I think, and a few months ago the idea was doing the round again.

Google for Network Computer and Larry Ellison, you would get many many hits.

Having dug around a little on your suggestion it appears as though the core idea is pretty similar to the Google proposal, see here for an editorial overview. In essence the idea harks back to what I was saying to Bill earlier that this ideology moves more towards the use of applications over a network without the discrepancy of what system the end user is using - although there will still be differences in the server side implementations, this will be completely transparent to the end user.

Thanks for the pointers abut Larry Ellison, Oracle and Network Computers.

Dave

 

Hi,

Running apps on the net might have some hidden problems. I can recall some time ago when the stste university I worked for was still in love with its mainframe computer (the kind that had two IBM techs on duty 24/7). About the time it took 45 minutes to get your terminal logged in, the powers that were got the message about desktop computers being more time and cost efficient.

How many zillion people can hit the Google site and be able to do their word processing before things get a bit tedious?

Also, home computers can't quite become thin clients and still be able to do anything that isn't run on the Google/whoever server. It seems kind of limiting - ya gotta pay for the hardware and most likely for the application usage. But if they aren't thin clients, then the OS issue (and browser) come back to life.

Maybe I'm missing the point here. I can see only one advantage to it, which would be to slow down the obsolence of hardware because computer evolution would essentially freeze. I'm grumpy about this because I lost another piece of hardware from my system because I have gone up to XP, and my scanner does not have a driver that works any longer. I'm tired of trashing otherwise functional peripherals.

 

Hi,

Running apps on the net might have some hidden problems. I can recall some time ago when the stste university I worked for was still in love with its mainframe computer (the kind that had two IBM techs on duty 24/7). About the time it took 45 minutes to get your terminal logged in, the powers that were got the message about desktop computers being more time and cost efficient.

How many zillion people can hit the Google site and be able to do their word processing before things get a bit tedious?

Also, home computers can't quite become thin clients and still be able to do anything that isn't run on the Google/whoever server. It seems kind of limiting - ya gotta pay for the hardware and most likely for the application usage. But if they aren't thin clients, then the OS issue (and browser) come back to life.

Maybe I'm missing the point here. I can see only one advantage to it, which would be to slow down the obsolence of hardware because computer evolution would essentially freeze. I'm grumpy about this because I lost another piece of hardware from my system because I have gone up to XP, and my scanner does not have a driver that works any longer. I'm tired of trashing otherwise functional peripherals.

This is an interesting point Bill. When I was at University we ran all applications off a central mainframe over a network, and to be fair it was quite an experience. It wasn't helped by the fact that the client side PCs were pretty ancient (pretty shocking for the Electrical Engineering department at a Russell Group University!). Like you point out, it took forever to log in (this was probably more to do with the bloated client side software that administered such tasks) however once logged in running general applications was not too bad; for example running a word processor, spreadsheets and a web browser. The main problem arose when trying to do something which required a reasonable level of computing power, for example: we did several projects involving hardware designs on FPGAs, designing the system using VHDL was no problem, however when it came to testing the design using a testbench on student-crippled simulation software the shortcomings of the networked applications became apparent. It would take a complete overnight simulation to generate 20ns of simulation results if we were lucky, and the throughput of these simulations would vary dramatically should there be a group of students performing similar simluations at the same time.

Having considered this I drew the following conclusions: For most people, the most they will ever do on a computer is write a couple of letters, browse the web, look at a couple of pictures from their digital camera and send an e-mail to their Aunte May. Yet they are sat there using a PC with a 3GHz processor, 1024 MB RAM and advanced graphics card with accelerator, the poor dual-core Pentium 4 spends 99.9% of its life executing the fruitless Windows System Idle Process. This idea of distributed computing would be ideal for these people, where they can have minimum spec computers that will allow them to do just what they want with the excesses of modern PCs. People who do more intense computing will probably still need a high-spec client side PC to allow them to do what they want.

I think the Google idea is just to provide the basic computing necessities to the average user I have described above, and this could be as many as half of those that use a PC.

Dave

 

Hi Dave,

You're no doubt about the computational needs of most people. I'll jump on the net app bandwagon as soon as I cna be convinced that logging onto the app server won't kick off a 25 minute commercial cycle, with pop-ups appearing every time you pause typing for 10 seconds.

Of course, there will be a "premium" service that eliminates all that, but it will be costly.

The way it works at universities - the administration gets all the good stuff. They get every new computer and PDA known to man. All departments get to spend limited bugetary funds to upgrade, which means only the dept. head has new stuff. The rest of the faculty can keep on with the old 5150, or spend their research grant on computers.

We got stuck under Campus Computing due to some inconpetence at the grad school office. Going to meetings was just wonderful - all the admins were busy showing off their new tech toys. We had to build our own computers from parts.

 

Hi Dave,

You're no doubt about the computational needs of most people. I'll jump on the net app bandwagon as soon as I cna be convinced that logging onto the app server won't kick off a 25 minute commercial cycle, with pop-ups appearing every time you pause typing for 10 seconds.

Of course, there will be a "premium" service that eliminates all that, but it will be costly.

The way it works at universities - the administration gets all the good stuff. They get every new computer and PDA known to man. All departments get to spend limited bugetary funds to upgrade, which means only the dept. head has new stuff. The rest of the faculty can keep on with the old 5150, or spend their research grant on computers.

We got stuck under Campus Computing due to some inconpetence at the grad school office. Going to meetings was just wonderful - all the admins were busy showing off their new tech toys. We had to build our own computers from parts.

Hi Bill, I almost missed this one!

Again you make a good point regarding the commercial aspects of such a proposition, whereas I can only talk about it from a technical point of view. It will certainly require a certain business model to make it feasible to finance all the servers that would be required to support such a service/OS alternative. I think in this day people are less forgiving of websites/webportals/webservices that bombard them with pop-ups and adverts. All major web browsers now have a pop-up blocker capbable of block 99.9% of pop-ups, and many browsers now have the functionality of an Ad/Content blocker to filter out most, if not all, of the embedded adverts in web pages.

The idea of a premium service has been in force for quite some time, look at the premium Yahoo! e-mail service that adds additional extras onto ad-free e-mail and 2GB storage space, including unlimited music streaming from Launch etc. For the sum of around $12 (this is how much it cost last time I checked) this seems pretty reasonable. So how much would they charge for access to an on-line suite of Office applications? And how much would this cost against what you would pay for MS Office? (for the sake of arguement we shall disregard the effect of the free OpenOffice.org). I don't have the answers to these qusetions, but ultimately in this game, its a buyers market.

As for the Univeristy supply chain, that has always been the case. Relaxing in plush exec-chairs with the latest 21" flat-panel monitor, all-singing all-dancing PCs and PDAs, all for sending a few e-mails through Outlook and writing the odd document. Whereas the poor engineer is sat there with an old P3 chugging away simulating some project!

Dave

 

I have moved the above topics from the original topic Linux, since this discussion have diverged enough for it to warrant its own thread.

Any questions or objections, please PM me. Thanks.

Dave