To precis the story: - popular legend has it that Billy the Kid was buried in place A - some guy said "I'm Billy the Kid" before he died, and he's buried in place B - another guy said "I'm Billy the Kid" before he died, and he's buried in place C
Surely places A, B and C are all benefitting from tourist dollars out of the pockets of people wanting to see the "real" Billy the Kid's hole in the ground.
Why on earth would the officials in place A want to (a) disturb the remains of 4 people (3 Kids and mum), (b) potentially wreck the tourism incomes of places B & C, and (c) potentially wreck the tourism income of their own town? If (c) happens, then I figure they won't be re-elected and will probably need to find a new place to live.
Surely the commonsense thing for guys in towns A, B & C is to leave things exactly as they are, and try to fuel the mystery of BtK's final resting place as part of his mystique.
"They think they've got him somewhere else, but old Mabel up the road remembers his grandson who lived just next door and he was a helluva shot with a gun, just like ol' Billy. Killed 150 buffalo in an afternoon with his bare hands, he did, and ate 'em all for dinner that night; never seen a buffalo in these parts since. You can see the family resemblance in this here photo, and you can get your own copy for only $5"
My understanding is that IBM currently recommends either SuSE or Redhat for its Linux customers, depending on where the customer is based. Given that lots of "big" customers have small offices in the boondocks, what are they going to recommend?
Small site typically equates to "we want it cheap, we want it reliable and we want it now. Even though we're part of a big company, head office says we have to keep our costs very low. If we don't we shut up shop". Once you add up lots of small sites, they actually carry a bit of clout in a large organization; you'd better be able to deliver a solution that fits their needs if you want to retain that customer. Quite often, a small site exists solely to service one big customer; global HQ wants to keep that small site happy.
Non-enterprise RedHat fit the bill perfectly for small sites, but SuSE might be too expensive given the lack of a download-only release. I'd assume IBM was hoping Fedora might be a good substitute for non-enterprise RedHat, but if not, which way will they turn?
I know about copyright, but that doesn't prevent SCO from claiming something as theirs even if it's not.
What I want to know is whether there's something in the US legal system that can stop SCO just stringing the court out ad infinitum with bogus ownership claims on file after file. All SCO's execs need is another couple of quarters and they win, regardless of what happens to SCO afterwards.
Is there something in the US legal system that actively prevents SCO from claiming every single file in the kernel as their own? I mean, if they're claiming a huge chunk of the kernel anyway, why not claim it all and be done with it?
Presumably the court would have to investigate every single file to establish the veracity of SCO's claims. Can you imagine weeding through every single.h file, many of which are going to look identical on any Unix system, and trying to establish the one true inventor of the file? I'd bet the original creator of a lot of those files is lost in time.
SCO would claim ownership, nobody could dispute it, and SCO could conceivably table some obscure document saying they created blah.h on some vaguely appropriate date. It doesn't have to be truthful - it can be manufactured just like all their other claims. If someone eventually comes forward and says "No, I created blah.h and here's proof", SCO would say "Fine, now let's look at blah2.h...".
Meanwhile, time would pass & SCO execs would be that much closer to their bonus payouts for keeping the stock price going upwards.
Other than the cost of doing things in this very long drawn out fashion, and I suspect SCO could find external funding for their legal costs from one or two sources in order to keep the anti-Linux FUD alive, is there anything in the US legal system preventing them from doing this?
I'm thinking all SCO's execs need to do is waste the court's time until they've got their 4th quarter of stock price rises, then bail out with big final paycheques and leave the remaining investors with whatever's left.
> the music industry alleges the pirated music cost > it at least $60 million
That's one f*ck of a lot of Kylie!
Let's do a bit of maths on this. A CD in Australia costs around $20-25. Let's round this up to $30, to give ARIA the benefit of the doubt.
An average CD contains about 10 tracks.
I'm going to assume that ARIA used something resembling base-10 mathematics... $60 mill equates to 2 million CDs, or 20 million tracks worth of downloads.
That's one track for every person in Australia.
Let's further assume that each track was a 3Mb MP3 file, which is probably a bit on the low side. The 20 million tracks that were downloaded works out to about 60Tb of data.
Are we supposed to believe that these guys, using a site running from a suburban bedroom, managed to share 60Tb of data? **Maybe** ARIA's lawyer is assuming that each track that was downloaded from this site was copied to another 10 sites, and from each of these to another 10,... - if so, that's hardly the fault of Mr Ng and his cohorts.
Does anyone have any more info on this case? Preferably, something a bit more credible?
If software patents are going to exist, then it makes sense to patent things you invent in order to protect yourself. If not, and your invention is worth anything significant, some other dweeb will come along, patent the thing you invented, then sue you for using it. That becomes a problem if said dweeb has deep pockets (e.g. Eolas, Microsoft,...) and you don't.
If you own patents on anything, there's nothing stopping you from licencing these patents for $0 to anyone else in the world. In that light, I'd rather Mozilla *did* try to patent the technology, then licence it out to the rest of the world free of charge. Better yet, get the patent and donate it to the EFF. That way I'd be reasonably sure popup blocking would remain free for anyone to implement/use however they saw fit.
Of course, if software patents didn't exist, the whole discussion would be moot and the world would be a much happier place IMHO.
An entirely new approach to running a business isn't cool? Did we learn nothing from the dot-bomb experience? Consider this:
1. Alienate potential customers, existing customers and existing resellers of your products. 2. Piss off biggest computer company in the world 3. Piss off regulatory groups in many countries in the world 4. Make unsubstantiated demands for money from complete strangers. Provide no mechanism whereby said strangers can pay the requested money 5. Threaten all and sundry with lawsuits. Specifically refuse to tell anyone the basis for these lawsuits, with the explanation that doing so would allow people to stop "breaking the law" which would be unacceptable 6. Issue verbal indemnities to potential lawsuit recipients, based on constantly changing criteria. Ensure nobody can tell whether they're actually indemnified or not, which actually isn't that big a deal as nobody knows what the basis of the lawsuits is anyway (refer previous point) 7. Continue distributing for free that which you proclaim to be illegal to give away. Distribute it under a licence that specifically says "it's OK to give this away" 8. Sell product in blatant violation of the rights of the owners of the product. Simultaneously complain that others are doing the exact same thing with "your" product
If the Iraqi Information Minister was cool - and who among us thought he wasn't THE man for 2003? - then this is also cool.
> "For the computer-illiterate home user, Windows is > fine"
> No, they're the worst ones to have it. They are > the ones whose box gets taken over and use to > spread worms, and DoS attacks.
Yes, yes, yes! Despite rumors to the contrary, there are plenty of people who use PCs at home, don't know a lot about them, and who don't have a requirement to play the latest and greatest games.
Their apps typically consist of: - a Web browser - email client, which may be a Web browser that they use for Web mail - something to write documents with (could be MS Word, could be MS Works, could be Wordpad, could be Notepad,...) - maybe a spreadsheet, if they're really advanced
If they're really high-tech, they might also have: - something to download pictures from a digital camera - a scanner to scan old pictures - graphic software to manipulate pictures
A Linux distribution that had *just* these features would be ideal for this class of user. They don't download virus updates, or configure firewall software - a well-built Linux distribution would eliminate the need to do these things.
If this was on the market now, I'd get it and put it on my parents' two PCs as quickly as I could.
> Aren't most COBOL applications deployed on big > iron?
That's been my experience too. Most COBOL work being done today seems to be primarily presenting existing data in different formats for consumption by Unix or Windows based systems.
There's no way this COBOL code is going to be migrated off the mainframe, for several reasons: - it's being written and maintained by mainframe COBOL coders. They have no interest whatsoever in saying it's feasible/viable/cost-effective/... in porting this code to another platform, because doing so would probably put them out of a job - this code tends to be built on top of old code, which is built on top of older code, which is built on top of... It's very rare to find someone writing COBOL code from scratch; it's all based on changing proven existing code and thus carries a relatively tiny risk of problems occurring after deployment. Furthermore, it's quite common to take the output of one piece of COBOL code, and use that as the input for your new set of COBOL code. Add a few generations of this, and suddenly you're dealing with an enormous mass of COBOL code, possibly largely undocumented and written by guys who retired years ago, that you have to port across to get a single piece of functionality working. It's very difficult to even write test cases for this legacy code, since its original purpose may be totally forgotten and the only reason it still exists is to support all the newer code that's been layered on top of it. This is a big factor that tends to be conveniently forgotten by those who think a move off the mainframe to Windows or Unix is simply a matter of time. A change to the.NET platform, or any other platform for that matter, is simply not feasible for the vast majority of mainframe shops on this basis alone. - as this code is primarily involved in presenting data in different formats, this work is best done as close to the database as possible. You don't want to be sending a big wad of data to another system, only to have 90% of it get thrown away and the remaining 10% formatted and consumed. It's generally cheaper to do this on the mainframe - in the mainframe environment,.NET is often seen as a high risk option. Remember, CICS and MVS have been around for decades, and they're known to work close enough to perfectly. For the type of work COBOL is currently being used for, reliability is absolutely top priority; you don't want your huge transaction processing system to go off the air for even a few seconds while your Windows systems take an outage for patches to be applied.
That's my take on things, but I'd be very interested to hear from anyone who sees mainframe COBOL code being used to do anything different.
I'd say you're getting a signal that your business model is no longer valid. Regardless of whether you think it's fair or reasonable, your customers are telling you they don't want to see your advertising and now they don't have to.
It sounds very harsh, but it's time to adapt or die. You can say "naughty user" or "bad Symantec" all you like, and maybe even get a lot of people to agree with you, but that isn't going to solve your problem.
Actually there's another choice: you can try to leverage the legal system to extend the life of your business model. It may even work in these not-so-enlightened times.
1. Create new release 2. Install shiny new distributed agents on Web servers 3. Announce new release on Slashdot 4. Examine performance of Web server agents under extreme load. Saves all that expensive load testing time and expense... 5. Announce new "stress tested version" shortly afterwards
Love to argue with you, but I can't: configuring Mandrake so that you can watch DVDs is a huge pain.
Even if you're savvy enough to track down the PLF site, it's still ugly getting all the bits working together.
Best way I've found is to just forget Mandrake and use Movix or Movix2 for watching DVDs, but that's hardly valid for my "typical home user". Hopefully the necessary bits of software will be integrated into the Mandrake distribution sooner rather than later.
From the article, it seems that he's making a fairly common misconception: he's saying it's difficult to install Linux.
How many "normal" people out there have ever installed Windows? You buy a PC; it has Windows pre-installed on it; the end. I'm not sure, but I suspect MS sales figures would support this; the vast majority of Windows sales are bundled with new hardware.
If he thinks installing Linux for home use is hard, try installing Windows for home use. First you install from the XP CD, then you'd better call MS to get it activated. Then you start installing all those patches from Windows Update. Then you start installing your apps - one at a time, and you'd better have all those code numbers and activation keys at your fingertips. Don't forget to call Symantec to register their products, and Quicken too. Don't forget to track down the driver CDs for those obscure bits of hardware - in this sense, "obscure" can mean things like digital cameras, scanners, etc. that are actually pretty common in home PCs.
In my experience, the typical home user installing a Windows PC from scratch takes 1-2 *days* to get it done, and that's 1-2 days of dedicated time spent largely sitting in front of the computer. Remember we're talking typical home users here, not corporate desktops or home machines owned by techos.
Now look at the Linux way of doing things. You get a Linux distribution from somewhere, and this may be a challenge if you don't know where to start looking. You power up the PC, put in the first distribution CD, and off you go. The installers for all the major Linux distributions are now pretty well comparable with Windows in terms of ease-of-use, although driver support is a bit more challenging.
You pick what sorts of apps you want (e.g. word processor, spreadsheet,...) then say "Go". After a while, you'll be prompted to put in the next CD (unless you're installing off DVD), then the next one. Once that's done, you reboot and you're done - there's very little need for a home user to install patches to things like Mozilla, KDE because they simply don't need them.
Unlike Windows Update patches, most patches to "Linux software" is to add functionality or protect against obscure buffer overflows - again I'm talking about "typical home user" stuff. Most of it just isn't needed.
I just can't see how installing Linux is even remotely as difficult as installing Windows these days. Typical time to install Linux, from scratch, for a new home user is a few hours - admittedly most of that time is head-scratching time, but it's still a whole lot less than 1-2 days of typing in codewords and swapping CDs on the Windows platform.
Hell, if you want to really reduce the time just get the home user to boot up Knoppix. Plug in a USB memory card and they can back up everything to it. There's your install done, in a couple of minutes (and that includes the trip to the shop to buy the memory stick).
Even though I work with a bunch of MS consultants who all carry these, I'm yet to see the killer use for tablet PCs that actually makes them more useful than a laptop PC. While laptop vs. tablet remains largely a matter of personal preference, tablets just won't sell.
I can see several potential vertical markets for tablet PCs, but they lack the following: - a killer app or suite of apps for any one vertical market - a distinctly "better" interface than laptop PCs for any market - a new group of users; ones that wouldn't use laptop PCs, but would use one of these - a cost point that makes them a worthwhile investment. In particular, the Windows OS+tablet interface and the hardware requirements to run them blows the cost out too much - a much simpler interface. Current tablets are too complex for current non-PC users to use; manufacturers should be looking to *remove* stuff from tablets to make them simpler to use and cheaper to purchase in bulk. Something like an X-terminal with local storage and the ability to sync to central servers makes more sense than an all-singing-dancing laptop-like thing; the only people likely to buy the current crop of tablets are current laptop owners
However...
If the price of the hardware came down, and the tablet was reasonably rugged, I could see some opportunities for schools to take these up. I'm talking primary/secondary schools, or K-12 in US-speak. Take out the price of the Windows licence, and they're suddenly a lot more attractive.
Most schools are having their budgets slashed, so IT spending is very low, but imagine taking a bunch of these and hooking them to something like an LTSP server: - teachers could use them to mark attendance; most current teachers are hopeless with a keyboard, and prefer writing. I'm inclined to think that a full-screen app that looks just like an attendance book, which has the names of all the students and check boxes next to their names, would go down very well with teachers, particularly if they could enter a "tick" or "cross" with a pen rather than typing stuff in or using a mouse. Laptops just aren't working out for most non-maths/science teachers who have them, in my experience - teachers grading assignments, particularly if they could take the tablets home, enter their data then do a seamless sync back to the central server - kids using them in test environments; imagine entering all your answers on the tablet, having the handwriting recognition neaten up the answers, then (for some subjects) getting your results at the end of the class. The marks could immediately go into a central database, and be exposed to parents over the Web; lots of possibilities here... - learning tools for specific, visual/factual subjects (e.g. geography; imagine all those maps coming to life when you point to them...)
In particular, using LTSP, there's no need for every kid to have one; they could be tied to a class or classroom rather than an individual kid. Maybe physically lock them to desks somehow, or use RFIDs to track their whereabouts - I don't know, ask a hardware guy...
I'd say this could be an interesting opportunity for a vertical market in education.
BTW, IANAT (...teacher), but I have a mother who's an English teacher, and who constantly complains about using laptops because they're just too complicated.
What's this "us" and "you" business? Here in Australia, we're still having the virtues of DMCA-like laws spruiked to our politicians - we've got at least another few months before they're law.
The "sheriff of Asia" still has to go down this particular stupid path.
The trick with Knoppix is that you actually boot off the Knoppix CD - unless you've configured the BIOS to prevent this, and password protected it, and there's no way of resetting the BIOS password, you're powerless to stop people booting Knoppix unless you physically remove the CD drive from the box.
It's a security risk if you store any sensitive files on the user's system - no doubt about it. On the other hand, it's God-like when it comes to recovering data from non-booting Windows boxes.
Is the FSF (or whoever) able to set up such a mechanism? If I could collect a voucher, then donate the funds directly to the FSF, I'd do it straight away. I mean, the amounts on each voucher are probably small enough that they're not worth much to anyone individually, but a large collection of vouchers would be worthwhile.
I'd prefer my donated funds didn't go directly to a project or FOSS company (sorry guys!), but to a group charged with ensuring FOSS' continued survival and/or prosperity.
If there's a way that the FSF (or whoever) can invoice me for "software", such that these vouchers can be cashed, it'd be potentially a huge win.
As a proud iPod owner, I have to respond to this;->
The controls are pretty sensitive, but when you're carrying it around with you, you use the remote and disable the controls on the iPod itself. The remote's buttons aren't anywhere near as sensitive - you have to give them a decent push to use them.
The sound quality is winning "audiophile" recommendations; it's more than "just OK".
With respect to the high price tag, Apple got smart when they put a (token) address book and calendar on there. They're very close to useless, but the existence of these features qualifies the iPod as a tax deductible expense for lots of people. If you like, the "50% greater price tag" is being subsidised by the government. The lack of these features on other MP3 players means they don't qualify as being tax deductible, so they "cost more" as a result.
The addition of several 3rd party add-ons is also helping to push the iPod. Along with the expected car charges, battery kits etc., there's a FM transmitter (listen to your iPod through the car stereo without wiring it up!), an addon to dump your digital camera's card contents onto the disc and a few others that aren't of any interest to me personally. The 3rd party market sets the iPod apart from the rest; these features simply aren't available on other MP3 players.
I have to agree about the 8 hour battery life, though; it's not enough to commute to and from work and to use at work, which is what I'd like. Furthermore, since the life of the internal battery is limited to so many recharges, I suspect the iPod becomes a throwaway item after a couple of years when the batteries won't hold a charge any more.
You can download the Sharp #Develop IDE; in terms of functionality and robustness, it's sort of an early beta level Visual Studio equivalent, but it's free and certainly makes writing.NET code a lot more enjoyable than using Notepad.
I'm sure Eclipse's C# support is pretty good too; just never had the need to use it personally.
Too late - a patent lawyer here in Australia successfully patented the wheel a year or two ago. He did it to expose the stupidity in the current patent system; not to collect royalties (at least as far as I know...)
Slashdot Mirror
To precis the story:
- popular legend has it that Billy the Kid was buried in place A
- some guy said "I'm Billy the Kid" before he died, and he's buried in place B
- another guy said "I'm Billy the Kid" before he died, and he's buried in place C
Surely places A, B and C are all benefitting from tourist dollars out of the pockets of people wanting to see the "real" Billy the Kid's hole in the ground.
Why on earth would the officials in place A want to (a) disturb the remains of 4 people (3 Kids and mum), (b) potentially wreck the tourism incomes of places B & C, and (c) potentially wreck the tourism income of their own town? If (c) happens, then I figure they won't be re-elected and will probably need to find a new place to live.
Surely the commonsense thing for guys in towns A, B & C is to leave things exactly as they are, and try to fuel the mystery of BtK's final resting place as part of his mystique.
"They think they've got him somewhere else, but old Mabel up the road remembers his grandson who lived just next door and he was a helluva shot with a gun, just like ol' Billy. Killed 150 buffalo in an afternoon with his bare hands, he did, and ate 'em all for dinner that night; never seen a buffalo in these parts since. You can see the family resemblance in this here photo, and you can get your own copy for only $5"
My understanding is that IBM currently recommends either SuSE or Redhat for its Linux customers, depending on where the customer is based. Given that lots of "big" customers have small offices in the boondocks, what are they going to recommend?
Small site typically equates to "we want it cheap, we want it reliable and we want it now. Even though we're part of a big company, head office says we have to keep our costs very low. If we don't we shut up shop". Once you add up lots of small sites, they actually carry a bit of clout in a large organization; you'd better be able to deliver a solution that fits their needs if you want to retain that customer. Quite often, a small site exists solely to service one big customer; global HQ wants to keep that small site happy.
Non-enterprise RedHat fit the bill perfectly for small sites, but SuSE might be too expensive given the lack of a download-only release. I'd assume IBM was hoping Fedora might be a good substitute for non-enterprise RedHat, but if not, which way will they turn?
Where's the bits that you insert in mouth and anus?
I know about copyright, but that doesn't prevent SCO from claiming something as theirs even if it's not.
What I want to know is whether there's something in the US legal system that can stop SCO just stringing the court out ad infinitum with bogus ownership claims on file after file. All SCO's execs need is another couple of quarters and they win, regardless of what happens to SCO afterwards.
Is there something in the US legal system that actively prevents SCO from claiming every single file in the kernel as their own? I mean, if they're claiming a huge chunk of the kernel anyway, why not claim it all and be done with it?
.h file, many of which are going to look identical on any Unix system, and trying to establish the one true inventor of the file? I'd bet the original creator of a lot of those files is lost in time.
Presumably the court would have to investigate every single file to establish the veracity of SCO's claims. Can you imagine weeding through every single
SCO would claim ownership, nobody could dispute it, and SCO could conceivably table some obscure document saying they created blah.h on some vaguely appropriate date. It doesn't have to be truthful - it can be manufactured just like all their other claims. If someone eventually comes forward and says "No, I created blah.h and here's proof", SCO would say "Fine, now let's look at blah2.h...".
Meanwhile, time would pass & SCO execs would be that much closer to their bonus payouts for keeping the stock price going upwards.
Other than the cost of doing things in this very long drawn out fashion, and I suspect SCO could find external funding for their legal costs from one or two sources in order to keep the anti-Linux FUD alive, is there anything in the US legal system preventing them from doing this?
I'm thinking all SCO's execs need to do is waste the court's time until they've got their 4th quarter of stock price rises, then bail out with big final paycheques and leave the remaining investors with whatever's left.
> the music industry alleges the pirated music cost
... - if so, that's hardly the fault of Mr Ng and his cohorts.
> it at least $60 million
That's one f*ck of a lot of Kylie!
Let's do a bit of maths on this. A CD in Australia costs around $20-25. Let's round this up to $30, to give ARIA the benefit of the doubt.
An average CD contains about 10 tracks.
I'm going to assume that ARIA used something resembling base-10 mathematics... $60 mill equates to 2 million CDs, or 20 million tracks worth of downloads.
That's one track for every person in Australia.
Let's further assume that each track was a 3Mb MP3 file, which is probably a bit on the low side. The 20 million tracks that were downloaded works out to about 60Tb of data.
Are we supposed to believe that these guys, using a site running from a suburban bedroom, managed to share 60Tb of data? **Maybe** ARIA's lawyer is assuming that each track that was downloaded from this site was copied to another 10 sites, and from each of these to another 10,
Does anyone have any more info on this case? Preferably, something a bit more credible?
If software patents are going to exist, then it makes sense to patent things you invent in order to protect yourself. If not, and your invention is worth anything significant, some other dweeb will come along, patent the thing you invented, then sue you for using it. That becomes a problem if said dweeb has deep pockets (e.g. Eolas, Microsoft, ...) and you don't.
If you own patents on anything, there's nothing stopping you from licencing these patents for $0 to anyone else in the world. In that light, I'd rather Mozilla *did* try to patent the technology, then licence it out to the rest of the world free of charge. Better yet, get the patent and donate it to the EFF. That way I'd be reasonably sure popup blocking would remain free for anyone to implement/use however they saw fit.
Of course, if software patents didn't exist, the whole discussion would be moot and the world would be a much happier place IMHO.
An entirely new approach to running a business isn't cool? Did we learn nothing from the dot-bomb experience? Consider this:
1. Alienate potential customers, existing customers and existing resellers of your products.
2. Piss off biggest computer company in the world
3. Piss off regulatory groups in many countries in the world
4. Make unsubstantiated demands for money from complete strangers. Provide no mechanism whereby said strangers can pay the requested money
5. Threaten all and sundry with lawsuits. Specifically refuse to tell anyone the basis for these lawsuits, with the explanation that doing so would allow people to stop "breaking the law" which would be unacceptable
6. Issue verbal indemnities to potential lawsuit recipients, based on constantly changing criteria. Ensure nobody can tell whether they're actually indemnified or not, which actually isn't that big a deal as nobody knows what the basis of the lawsuits is anyway (refer previous point)
7. Continue distributing for free that which you proclaim to be illegal to give away. Distribute it under a licence that specifically says "it's OK to give this away"
8. Sell product in blatant violation of the rights of the owners of the product. Simultaneously complain that others are doing the exact same thing with "your" product
If the Iraqi Information Minister was cool - and who among us thought he wasn't THE man for 2003? - then this is also cool.
> "For the computer-illiterate home user, Windows is
...)
> fine"
> No, they're the worst ones to have it. They are
> the ones whose box gets taken over and use to
> spread worms, and DoS attacks.
Yes, yes, yes! Despite rumors to the contrary, there are plenty of people who use PCs at home, don't know a lot about them, and who don't have a requirement to play the latest and greatest games.
Their apps typically consist of:
- a Web browser
- email client, which may be a Web browser that they use for Web mail
- something to write documents with (could be MS Word, could be MS Works, could be Wordpad, could be Notepad,
- maybe a spreadsheet, if they're really advanced
If they're really high-tech, they might also have:
- something to download pictures from a digital camera
- a scanner to scan old pictures
- graphic software to manipulate pictures
A Linux distribution that had *just* these features would be ideal for this class of user. They don't download virus updates, or configure firewall software - a well-built Linux distribution would eliminate the need to do these things.
If this was on the market now, I'd get it and put it on my parents' two PCs as quickly as I could.
> Aren't most COBOL applications deployed on big
... It's very rare to find someone writing COBOL code from scratch; it's all based on changing proven existing code and thus carries a relatively tiny risk of problems occurring after deployment. Furthermore, it's quite common to take the output of one piece of COBOL code, and use that as the input for your new set of COBOL code. Add a few generations of this, and suddenly you're dealing with an enormous mass of COBOL code, possibly largely undocumented and written by guys who retired years ago, that you have to port across to get a single piece of functionality working. It's very difficult to even write test cases for this legacy code, since its original purpose may be totally forgotten and the only reason it still exists is to support all the newer code that's been layered on top of it. This is a big factor that tends to be conveniently forgotten by those who think a move off the mainframe to Windows or Unix is simply a matter of time. A change to the .NET platform, or any other platform for that matter, is simply not feasible for the vast majority of mainframe shops on this basis alone. .NET is often seen as a high risk option. Remember, CICS and MVS have been around for decades, and they're known to work close enough to perfectly. For the type of work COBOL is currently being used for, reliability is absolutely top priority; you don't want your huge transaction processing system to go off the air for even a few seconds while your Windows systems take an outage for patches to be applied.
> iron?
That's been my experience too. Most COBOL work being done today seems to be primarily presenting existing data in different formats for consumption by Unix or Windows based systems.
There's no way this COBOL code is going to be migrated off the mainframe, for several reasons:
- it's being written and maintained by mainframe COBOL coders. They have no interest whatsoever in saying it's feasible/viable/cost-effective/... in porting this code to another platform, because doing so would probably put them out of a job
- this code tends to be built on top of old code, which is built on top of older code, which is built on top of
- as this code is primarily involved in presenting data in different formats, this work is best done as close to the database as possible. You don't want to be sending a big wad of data to another system, only to have 90% of it get thrown away and the remaining 10% formatted and consumed. It's generally cheaper to do this on the mainframe
- in the mainframe environment,
That's my take on things, but I'd be very interested to hear from anyone who sees mainframe COBOL code being used to do anything different.
I'd say you're getting a signal that your business model is no longer valid. Regardless of whether you think it's fair or reasonable, your customers are telling you they don't want to see your advertising and now they don't have to.
It sounds very harsh, but it's time to adapt or die. You can say "naughty user" or "bad Symantec" all you like, and maybe even get a lot of people to agree with you, but that isn't going to solve your problem.
Actually there's another choice: you can try to leverage the legal system to extend the life of your business model. It may even work in these not-so-enlightened times.
> a public that thinks animated cursors on the web
> are the dog's back wheels
"The dog's back wheels" is a truly inspired turn of phrase. If I had mod points at the moment, you'd have one right now.
1. Create new release
2. Install shiny new distributed agents on Web servers
3. Announce new release on Slashdot
4. Examine performance of Web server agents under extreme load. Saves all that expensive load testing time and expense...
5. Announce new "stress tested version" shortly afterwards
reading this who thought it might be intended as satire?
Love to argue with you, but I can't: configuring Mandrake so that you can watch DVDs is a huge pain.
Even if you're savvy enough to track down the PLF site, it's still ugly getting all the bits working together.
Best way I've found is to just forget Mandrake and use Movix or Movix2 for watching DVDs, but that's hardly valid for my "typical home user". Hopefully the necessary bits of software will be integrated into the Mandrake distribution sooner rather than later.
From the article, it seems that he's making a fairly common misconception: he's saying it's difficult to install Linux.
...) then say "Go". After a while, you'll be prompted to put in the next CD (unless you're installing off DVD), then the next one. Once that's done, you reboot and you're done - there's very little need for a home user to install patches to things like Mozilla, KDE because they simply don't need them.
How many "normal" people out there have ever installed Windows? You buy a PC; it has Windows pre-installed on it; the end. I'm not sure, but I suspect MS sales figures would support this; the vast majority of Windows sales are bundled with new hardware.
If he thinks installing Linux for home use is hard, try installing Windows for home use. First you install from the XP CD, then you'd better call MS to get it activated. Then you start installing all those patches from Windows Update. Then you start installing your apps - one at a time, and you'd better have all those code numbers and activation keys at your fingertips. Don't forget to call Symantec to register their products, and Quicken too. Don't forget to track down the driver CDs for those obscure bits of hardware - in this sense, "obscure" can mean things like digital cameras, scanners, etc. that are actually pretty common in home PCs.
In my experience, the typical home user installing a Windows PC from scratch takes 1-2 *days* to get it done, and that's 1-2 days of dedicated time spent largely sitting in front of the computer. Remember we're talking typical home users here, not corporate desktops or home machines owned by techos.
Now look at the Linux way of doing things. You get a Linux distribution from somewhere, and this may be a challenge if you don't know where to start looking. You power up the PC, put in the first distribution CD, and off you go. The installers for all the major Linux distributions are now pretty well comparable with Windows in terms of ease-of-use, although driver support is a bit more challenging.
You pick what sorts of apps you want (e.g. word processor, spreadsheet,
Unlike Windows Update patches, most patches to "Linux software" is to add functionality or protect against obscure buffer overflows - again I'm talking about "typical home user" stuff. Most of it just isn't needed.
I just can't see how installing Linux is even remotely as difficult as installing Windows these days. Typical time to install Linux, from scratch, for a new home user is a few hours - admittedly most of that time is head-scratching time, but it's still a whole lot less than 1-2 days of typing in codewords and swapping CDs on the Windows platform.
Hell, if you want to really reduce the time just get the home user to boot up Knoppix. Plug in a USB memory card and they can back up everything to it. There's your install done, in a couple of minutes (and that includes the trip to the shop to buy the memory stick).
Even though I work with a bunch of MS consultants who all carry these, I'm yet to see the killer use for tablet PCs that actually makes them more useful than a laptop PC. While laptop vs. tablet remains largely a matter of personal preference, tablets just won't sell.
I can see several potential vertical markets for tablet PCs, but they lack the following:
- a killer app or suite of apps for any one vertical market
- a distinctly "better" interface than laptop PCs for any market
- a new group of users; ones that wouldn't use laptop PCs, but would use one of these
- a cost point that makes them a worthwhile investment. In particular, the Windows OS+tablet interface and the hardware requirements to run them blows the cost out too much
- a much simpler interface. Current tablets are too complex for current non-PC users to use; manufacturers should be looking to *remove* stuff from tablets to make them simpler to use and cheaper to purchase in bulk. Something like an X-terminal with local storage and the ability to sync to central servers makes more sense than an all-singing-dancing laptop-like thing; the only people likely to buy the current crop of tablets are current laptop owners
However...
If the price of the hardware came down, and the tablet was reasonably rugged, I could see some opportunities for schools to take these up. I'm talking primary/secondary schools, or K-12 in US-speak. Take out the price of the Windows licence, and they're suddenly a lot more attractive.
Most schools are having their budgets slashed, so IT spending is very low, but imagine taking a bunch of these and hooking them to something like an LTSP server:
- teachers could use them to mark attendance; most current teachers are hopeless with a keyboard, and prefer writing. I'm inclined to think that a full-screen app that looks just like an attendance book, which has the names of all the students and check boxes next to their names, would go down very well with teachers, particularly if they could enter a "tick" or "cross" with a pen rather than typing stuff in or using a mouse. Laptops just aren't working out for most non-maths/science teachers who have them, in my experience
- teachers grading assignments, particularly if they could take the tablets home, enter their data then do a seamless sync back to the central server
- kids using them in test environments; imagine entering all your answers on the tablet, having the handwriting recognition neaten up the answers, then (for some subjects) getting your results at the end of the class. The marks could immediately go into a central database, and be exposed to parents over the Web; lots of possibilities here...
- learning tools for specific, visual/factual subjects (e.g. geography; imagine all those maps coming to life when you point to them...)
In particular, using LTSP, there's no need for every kid to have one; they could be tied to a class or classroom rather than an individual kid. Maybe physically lock them to desks somehow, or use RFIDs to track their whereabouts - I don't know, ask a hardware guy...
I'd say this could be an interesting opportunity for a vertical market in education.
BTW, IANAT (...teacher), but I have a mother who's an English teacher, and who constantly complains about using laptops because they're just too complicated.
Patent it! ...And book the lawyers for 10-20 years in the future when someone finds a use for it that can pay off the research costs.
What's this "us" and "you" business? Here in Australia, we're still having the virtues of DMCA-like laws spruiked to our politicians - we've got at least another few months before they're law.
The "sheriff of Asia" still has to go down this particular stupid path.
> the newly-formed Joint Committee on Communications
> Security...
It seems that whenever the term "security" is part of the name of a government body in the US, something bad is about to happen.
The trick with Knoppix is that you actually boot off the Knoppix CD - unless you've configured the BIOS to prevent this, and password protected it, and there's no way of resetting the BIOS password, you're powerless to stop people booting Knoppix unless you physically remove the CD drive from the box.
It's a security risk if you store any sensitive files on the user's system - no doubt about it. On the other hand, it's God-like when it comes to recovering data from non-booting Windows boxes.
Is the FSF (or whoever) able to set up such a mechanism? If I could collect a voucher, then donate the funds directly to the FSF, I'd do it straight away. I mean, the amounts on each voucher are probably small enough that they're not worth much to anyone individually, but a large collection of vouchers would be worthwhile.
I'd prefer my donated funds didn't go directly to a project or FOSS company (sorry guys!), but to a group charged with ensuring FOSS' continued survival and/or prosperity.
If there's a way that the FSF (or whoever) can invoice me for "software", such that these vouchers can be cashed, it'd be potentially a huge win.
As a proud iPod owner, I have to respond to this ;->
The controls are pretty sensitive, but when you're carrying it around with you, you use the remote and disable the controls on the iPod itself. The remote's buttons aren't anywhere near as sensitive - you have to give them a decent push to use them.
The sound quality is winning "audiophile" recommendations; it's more than "just OK".
With respect to the high price tag, Apple got smart when they put a (token) address book and calendar on there. They're very close to useless, but the existence of these features qualifies the iPod as a tax deductible expense for lots of people. If you like, the "50% greater price tag" is being subsidised by the government. The lack of these features on other MP3 players means they don't qualify as being tax deductible, so they "cost more" as a result.
The addition of several 3rd party add-ons is also helping to push the iPod. Along with the expected car charges, battery kits etc., there's a FM transmitter (listen to your iPod through the car stereo without wiring it up!), an addon to dump your digital camera's card contents onto the disc and a few others that aren't of any interest to me personally. The 3rd party market sets the iPod apart from the rest; these features simply aren't available on other MP3 players.
I have to agree about the 8 hour battery life, though; it's not enough to commute to and from work and to use at work, which is what I'd like. Furthermore, since the life of the internal battery is limited to so many recharges, I suspect the iPod becomes a throwaway item after a couple of years when the batteries won't hold a charge any more.
You can download the Sharp #Develop IDE; in terms of functionality and robustness, it's sort of an early beta level Visual Studio equivalent, but it's free and certainly makes writing .NET code a lot more enjoyable than using Notepad.
I'm sure Eclipse's C# support is pretty good too; just never had the need to use it personally.
Too late - a patent lawyer here in Australia successfully patented the wheel a year or two ago. He did it to expose the stupidity in the current patent system; not to collect royalties (at least as far as I know...)