This looks nice. I'll download it when the heat dies down a bit.
Now we've got the following live CDs: - Knoppix; perfect geek distro, just about every geek tool in one place. The Swiss Army knife distro - Mepis; excellent end-user distro, exactly the Linux distro for mum and dad - Morphix; customisable distro, put whatever you want on it
IMHO, the missing one is the "live server" CD. You boot from this and you get Linux servers, not workstation tools. It should have the following features: - stable/testing versions of all common servers (e.g. Apache, Postgres, MySQL, Zope, iptables, sshd, Postfix, courier-pop, Samba,...) - support for all the server-class hardware out there (e.g. RAID cards, SCSI/SATA discs, etc.) - when booted from CD, all servers are enabled but discs aren't mounted by default. You can have a play around with it, but you have to go out of your way to hurt yourself - when booted from disc, all servers are disabled but all discs are mounted. Login for the first time as root and you get asked "Which of the following services would you like to enable?"... - best-of-class GUI config tools for the servers for both Windows and Linux. Once you've installed the server, you then use the tools on the CD on a workstation to configure it - tools to migrate existing data from proprietary solutions (e.g. email and mailing lists from MS Exchange, ). These could run on client workstations rather than on the server, if required; obviously they wouldn't automate the migration, but anything that could reduce the workload would be worth considering - support for reading/writing configurations to USB key. Installs can run unattended using configs stored on the USB key. This would allow you to install fleets of identical servers (e.g. Web farm) quickly
I'm sure there's other requirements you could come up with, but this would let you quickly put an entire data centre together. MS in particular would find it hard to compete with this.
Why didn't the company involved strip or mutilate all the sensitive data their contractors were given to work with? There's absolutely no excuse to give contractors sensitive data for this type of work.
Hell, even replacing every letter with a random letter and every number with a random number in sensitive fields would probably be enough; rather than "Mrs Joan Smith", they could've been working with "Grc Meas Fesze" without any impact. Change all names, addresses and "comments" and that would just about do it.
Doing this would completely remove the possibility of this problem happening.
If I was "Mrs Joan Smith", I'd be suing the retainer of my personal data on this basis. No wonder they're still considering whether to tell the victims what has happened...
I'm interested in getting involved against this becoming law, and I'll contact you offline. I'm particularly concerned about the stupidity of extending copyright to meet the so-called requirements of large US-based corporations.
However, it's important to realise that Australian common law offers built-in protection against a lot of the worst of the DMCA. Common law exists to remove the need for laws to deal with highly-specific information. It's one of the reasons why we don't get the continual barrage of new legislation that the US seems to get to vote on every few weeks.
Common law pretty much equates to "common sense". As an example, AFAIK there have been no prosecutions for illegal recording of TV shows - that's because we acquired certain common law rights when we purchased a VCR and one of those is the ability to use the "record" button.
Australian common law also protects (IIRC) e.g. decompiling code for the purposes of creating interoperability - it's no coincidence that projects such as Samba were started in Australia, as they are legally protected via a common law right to "use something you've purchased in a reasonable manner" (OK, that's oversimplifying, but it's not far from truth).
AFAIK, common law would "trump" any DMCA-like law that was introduced. Common law is enshrined in the Australian Constitution, and that isn't likely to be tweaked to accomodate Johnny-come-lately stuff like DMCAs.
Common law is one of the reasons why there's no overwhelming push for a US-style Bill Of Rights in Australia; many things that would be covered by it are already covered by common law. Any introduction of a Bill Of Rights here would probably only create grey areas that don't currently exist. Although there's very little to actually prevent it, we also don't have governments subjected to massive and well-organised financial lobbying from interest groups (e.g. large companies); such lobbying in the US is primarily driven by a proposed transfer of rights from individuals to those interest groups, and common law protects us as individuals to the point where these lobby groups have no reason to exist here.
Oh, IANAL but I've dated a few... A useful reference on common law is at http://www.austlii.edu.au/au/special/alta/alta95/m ason.html
Carrots: - Offer better/more complete Thai language support in Win Lite. - Offer it at low cost, as we now know will happen
Sticks: - *Any* hardware upgrade results in needing to re-register with MS Thailand. More memory, a CPU upgrade; just keep re-registering every time. That way they can track usage that much better, and possibly get early warning of changes in usage trends. - Possibly remove functionality to the point where Win Lite is essentially useless in corporate environments. Limit access to file servers, or network printers, or something similar; maybe time bomb it so you have X days to upgrade to the more expensive version once you start using "advanced" features - Limit it to the point where it's essentially useless outside of Thailand. That shouldn't be hard given the uniqueness of Thai language
> One benefit to having email is the ability to post > information anonymously in order to avoid possible > repercussions.
The obvious, but possibly overly-simplistic, solution to this is to change the email programs we're using. Make them configurable to either of these settings: - I'll only accept email with "warranties". This would be the "normal" setting - I'll accept email from anyone
That would mean, if you send "anonymous email", there's no guarantee it's gonna be accepted at the other end. Hopefully, sufficient people would choose the first option to kill off the spam market.
Whenever an "I won't accept anonymous email" user connects to his/her mail server, the mail server could then ditch the unwanted email for that user.
As the parent points out, there are valid reasons to send anonymous email, but the recipients of those messages will normally know in advance that they'll be receiving anon email and can set themselves up to deal with it. I'm thinking crisis support groups, specific mailing lists and so forth.
*Maybe* this could work - I'm sure there's a few aspects that have escaped me at sparrow fart in the morning...
The ZDNet article points out that if all the "pirated" tracks in Australia were purchased for $A0.99, then the record companies would be $2b better off.
As of now, my understanding is that Apple sells tracks for $US0.99, and is in pretty close to a breakeven state for iTunes (this may have changed recently, as surely the sheer volume going through iTunes would cause them to move into profit at some point). Regardless, it seems that $US0.99 is pretty close to the breakeven point, and you'd assume the breakeven cost in Australia would be no lower than that given the population is so small - let's cut the record companies some slack and assume $US0.90 is the breakeven point for online music sales in Australia.
$A0.99 translates to $US0.76. Now, since it costs $US0.90 to provide a downloadable music track, Kazaa is actually *saving* the record companies $US0.14 per downloaded track. By my calculation, the 850,000 tracks downloaded via Kazaa haved saved Australian record companies $US119,000 in providing that service.
> I met one once at an Open Source conference and > when you ask him what he does he very plainly > states "I'm a spammer". The guy was a total > pariah....So you're implying that the other people at the Open Source conference weren't total pariahs, then?
Although I switched to Debian some time back, I still buy every 2nd or 3rd Mandrake release and keep it on the shelf. Of all the for-pay distributions, I like Mandrake's model the best and I think there'd be a big hole in the desktop market in particular if Mandrake was to disappear.
The reason I switched to Debian was Knoppix. The hardware detection is that good; the install time is tiny; the products on the CD are pretty close to the full set of what I want.
The reason I use Linux is the fact that, when I install it, it just works and keeps on working. My experience is that a Windows desktop has a working life of about 6 months; after that, you're faced with a box that runs slower and slower and more and more problems are occurring. The only way of fixing that is to reinstall it, which consumes about 2 days by the time you've installed every app, every service pack and found every serial number (note: we're talking home PCs here). I've got a Linux file server that's run essentially unchanged except for disc upgrades and security patches since 1998, and has never once given me a problem - that simply doesn't happen with Windows boxes.
Installing a Linux box is a breeze these days. From bare metal, I can get a Linux desktop box installed and fully working *with all applications* within about 1-2 hours. For servers, it's often less. If necessary, I can then kick off an apt-get upgrade, come back in an hour or so, reboot and the box is operational. From bare metal, unless you're installing from an image, it takes many more hours sitting in front of a Windows system to install apps and service packs. I hate that I have to keep clicking "OK" or typing in serial numbers, so I can't wander off and do something else while the box installs. If you've got e.g. Ghost, you're covered to some extent, although then you hit problems if your hardware config is significantly different from that used for the Ghost image.
I like the fact that my Linux boxes never have downtime due to viruses or other rubbish. Whether this is due to market share or not, I don't care - my desktop Linux box hasn't had downtime other than for hardware changes for years.
I like that I can pick up my personal config from a Linux box and migrate it to new hardware just by copying/home. I don't have to rebuild configs that were stored in registry keys on Window boxes and thus lost during a migration.
I wonder why Apple hasn't made more of this in their advertising. It seems that having possibly the highest-profile series of movies in many years put together using your gear would be worth telling people about.
If the article is accurate, it's a great example of working globally that a lot of Apple's potential customers might want to hear about.
It'd certainly attract more positive interest than those ridiculous "HP Invent" advertisements - they're just laughable. Every time I see a new one, I think "What the hell am I looking at?" which I suspect isn't the message HP wants to be getting across.
> What can you say or do in a forty minute long > meeting?
Actually quite a lot. I'm assuming they didn't spend the first 20 minutes talking about the latest sporting results or TV programs.
I imagine the Pres could well have heard about a bunch of FOSS stuff, and, rather than reading through various interpretations of the GPL, thought it might be worth having a chat to the guy that wrote it. If I understand it correctly, the role of the President in India is largely to "suggest directions" for the country to go in, and I could well see this meeting as being pivotal in any additional embracing of FOSS in India.
I'd love to know the types of conversations Bill Gates has with world leaders, which seem to occur on a fairly regular basis. It's not as though MS actually employs loads of people worldwide, or directly contributes any sizeable chunk of loot to non-US economies, which is what's gonna be of interest to foreign heads of state. Sure MS supplies a lot of infrastructure stuff to governments, but you don't see e.g. the president of Ford Motor Co or Xerox getting meetings with world leaders on any regular basis. I just can't see that the conversation would be anything more involved than "buy my stuff" or "build this infrastructure so my stuff can do more things". Exactly why this type of conversation would merit the time of a world leader escapes me, yet it happens often enough.
> I've never seen a middle manager promoted from a > technical job.
There was a time not so long ago when techos hit a form of glass ceiling in a company of any size. You could either sit at basically the same pay scale forever as a pure techo, slip sideways into consulting/marketing/sales or move into management. Strangely, the thinking was that the best/only person to manage techos was an ex-techo, so there was never any real shortage of those jobs around.
> The average middle manager wouldn't know a > Postfix/procmail e-mail system if it crawled out > of their ass, jumped up on their desk and did the > tap number from 42nd street.
Agreed.
However, the middle manager who runs those systems know they're spending a sh1tload on trapping viruses and storing spam - or he would if the people working for him told him so.
He'll also know that ditching that spam and not getting nailed by viruses every few weeks will save him money.
I've got several mates who moved from geekdom into middle management over the years, and they're doing these sorts of changes. They might be far enough removed these days to not know about Postfix/procmail, but they're tuned in enough to see that it can solve their problems once the basic concepts have been explained to them.
They're seeing their teams increasing in size, while those of less technical managers are shrinking. They're also getting pay raises for themselves and their teams. Funny how solving business problems will do that for you...
> Corporate middle management is not interested in > facts. They are not interested in improvement. > They are not interested in efficiency that is not > accomplished by either making people shovel shit > or firing people.
> Middle management seeks to maintain the status > quo, and to do nothing unless it is absolutely > necessary. Incompetence, bankruptcy, waste, > stupidity, anything is better than trying and > failing.
I'll call bullshit on this.
Corporate middle management (MM) now faces the repercussions of CIOs and senior management telling shareholders "we'll reduce the cost of IT by 20/30/50% in the next 12 months". Middle management then gets told "do this or die"; either they slash their costs by A LOT within 12 months or they're out the door. Whatever was the case in the past, MM is now *all* about efficiency.
A sizeable chunk of MM has worked out "Hmm, if we keep doing what we have been doing, we'll keep getting the same results, so now we have to try something different". In many cases, they don't yet know what "something different" is or should be, but they are on the lookout for something - anything - that means they won't be leading their team into the unemployment office in 12 months' time.
**Now** is the best possible time to go to these MM guys with your ideas.
I'll give you an example: want to put e.g. Postfix/procmail in front of Internet-facing MS Exchange servers and use it to (a) de-evil incoming email with evil HTML content such as @ signs in URLs, and (b) filter out email from known open relays? Collect some figures on how much time/money has been lost in your org fighting spam and the latest HTML-based email virus, drop those figures on your MM's desk along with the costs of implementing your solution. If you do it right, your MM will realise, if it's done right, it'll slash his costs hugely and maybe get him a few percent closer to keeping his (and your) job intact.
The trick is to present data that makes sense to your MM. Don't tell him "we'll block 13432 incoming spam messages per day"; tell him "we'll block 13432 incoming spam messages per day that cost us $2300 per day in storage costs. My solution will cost us $3000 to implement, so it's paid for itself by the 2nd day". He has to talk in terms of financial outcomes, because that's what his boss wants; if you want to get your ideas across, you have to do the same.
Many techos, and I've done this myself in the past, present their ideas in such a way that it comes across as "It'd be really cool if we did X, and there might even be some benefits for the company if we did it. We're not quite sure exactly what X will cost, or how long it's gonna take to do it, but we should do it anyway because my geek buddy did it and he's really smart". It only took about 300 rejections before I worked out that this approach never works unless your boss has a goatee...
> Although it sounds like a racket, I have been told > that professors don't get royalties for books sold > at their own universities. This is to prevent the > abuse that you just mentioned, which although > cynical, is not true.
I'd love to see some proof of this.
When I did Comp Science at Melbourne University many years ago, it was remarkable how: - for each class, there was exactly one recommended textbook - the author of that textbook was the lecturer - that textbook was only available at a single place, the MU Bookshop, and at a price that any sane person would call exorbitant even for an IT book - new editions of those books appeared fairly regularly, with just enough changes so the page/chapter numbering was thrown out compared to the previous edition - I'm fairly confident *none* of these textbooks were ever opened again once you'd finished that subject. They held very little of any use outside of passing that subject
> The United States Government has a duty to protect > the interest of it's own citizens. It does *not* > have the duty to ensure that Indians get wealthy.
The same argument works in reverse too. The Indian government has a duty to protect the interest of its citizens too, and I bet they'd be very interested in anything that could potentially derail their huge and growing IT businesses.
Consider: most US-based IT companies stopped being "US companies" a long time ago. Companies like Microsoft, Hewlett-Packard, etc are now global companies that started and are based in the US. They *don't* have to support their country of origin, or even their employees, as a "prime directive"; they're primarily answerable to their shareholders.
If Microsoft was to stop using Indian coders (disclaimer: I only assume they're doing so), then the Indian govt would go straight to MS India and tell them "give us back *our* jobs or we'll stop buying your product". MS India would then relay that to MS Redmond, who would then have to weigh up the potential loss in Indian sales vs. other alternatives. With the amount of dollars being talked about and the global nature of their customers and workforce, there's probably a good argument for MS to move out of the US - they could relocate their head office to some other country that didn't try to control where their workers came from.
Would the US government punish MS if they did this? Almost certainly, although that punishment would have to be based on not buying MS' products and the US govt is probably lagging behind govts elsewhere in the world in investigating the viability of non-MS software. Even though the US govt is the single biggest customer for MS, I'm not sure that the scope of punishment that would/could be dished out by the US govt would be sufficient to sway the decision.
If MS moved its base to e.g. India, what would the Indian govt do? Probably commit to buying more MS software, and probably allow MS to go into govt markets that they previously haven't had access to.
What would MS' current shareholders do? I bet they'd support it, on the basis that MS should be able to reduce their costs significantly and a loss of sales in the US should be compensated by additional sales in India (in the case where MS relocated to India).
I'm not saying that this is likely to happen, but I bet it's a scenario that the US govt and the large "US IT companies" have investigated fairly thoroughly.
> The biggest problems with knoppix are (1) it > uses kde instead of gnome and (2) it has its own > package structure that is incompatible with > debian. So apt-get dist-upgrade or even apt-get > upgrade will break everything. I've only had > success upgrading individual packages with apt.
I've done both apt-get upgrade and apt-get dist-upgrade several times (over a period of a few months) on my installed-to-hard-disc Knoppix box, and haven't had a problem with it.
I've also installed an extra zillion games via apt-get for my kids to play on the same box, and they work fine too.
If you're having problems with this, it might be worth reporting it to the www.knoppix.net. The PC I used is a grey box clone running an old Celeron 533 with no "tricky" hardware whatsoever; maybe you're hitting problems with the specific hardware you're using.
SCO's documents continue to gloss over several key points: - code that e.g. Andrew Tridgell writes for Samba cannot possibly be owned by SCO - Samba code is released under the GPL. The GPL says, in effect, "I'm allowing you to use this code under a certain set of conditions". If the GPL is invalid, then basic copyright law will be in effect - SCO distributes Samba code, presumably under the GPL. I'm guessing that SCO hasn't negotiated a separate agreement with the Samba guys to distribute their code under some non-GPL arrangement - if SCO succeeds in getting the GPL rendered invalid (and that is unlikely since, in this example, it's the Samba guys saying "I'm gonna give you extra rights to use this provided you stick to these conditions..." which is a very common approach to licencing), then SCO is simply breaking basic copyright law in distributing Samba
In other words, if the GPL is valid, then SCO has no case. If the GPL is invalid, then SCO is breaking the copyright of lots of individual copyright holders. Either way, they lose
I remembered hearing Fels wasn't entitled to use the title "Professor" on the media, and Google turned up http://www.abc.net.au/mediawatch/transcripts/s8367 77.htm which, although inconclusive, implies that he *is* a Professor and is entitled to use the honorific. The confusion seems to have come about due to some out-of-date documentation supplied by a university in error.
This: http://www-pso.adm.monash.edu.au/news/Story.asp?ID =514&SortType=6 seems to confirm it.
Apologies for listening to the media and believing what I hear. Anyone want to buy these SCO shares; I believe the price is headed for the moon!
It's a mystery how this "Professor" title got prepended to Fels' name. For all that he did a remarkable job while in charge of the ACCC, the "Professor" title is totally bogus - he can no more call himself "Professor" than an unqualified person can call their self "Doctor".
Having worked on both sides of the fence, I can tell you: - IT managers value frankness. Yep, even the ones who give an unpleasant reaction to frank information appreciate the input, in my experience - if you're gonna slam someone, you'd better be prepared to offer good justification for your opinions. "Fred is a dickhead" wouldn't get you a pleasant response; "Fred is a dickhead because..." conceivably might if you can give a good enough explanation
Norton Spam Filter 2004, now with haiku filtering! Guaranteed to filter 100% of spam, as long as the Internet doesn't resort to copyright infringement...
You know I really tried, but I just can't weave a SCO comment into this message...
Slashdot Mirror
This looks nice. I'll download it when the heat dies down a bit.
...)
Now we've got the following live CDs:
- Knoppix; perfect geek distro, just about every geek tool in one place. The Swiss Army knife distro
- Mepis; excellent end-user distro, exactly the Linux distro for mum and dad
- Morphix; customisable distro, put whatever you want on it
IMHO, the missing one is the "live server" CD. You boot from this and you get Linux servers, not workstation tools. It should have the following features:
- stable/testing versions of all common servers (e.g. Apache, Postgres, MySQL, Zope, iptables, sshd, Postfix, courier-pop, Samba,
- support for all the server-class hardware out there (e.g. RAID cards, SCSI/SATA discs, etc.)
- when booted from CD, all servers are enabled but discs aren't mounted by default. You can have a play around with it, but you have to go out of your way to hurt yourself
- when booted from disc, all servers are disabled but all discs are mounted. Login for the first time as root and you get asked "Which of the following services would you like to enable?"...
- best-of-class GUI config tools for the servers for both Windows and Linux. Once you've installed the server, you then use the tools on the CD on a workstation to configure it
- tools to migrate existing data from proprietary solutions (e.g. email and mailing lists from MS Exchange, ). These could run on client workstations rather than on the server, if required; obviously they wouldn't automate the migration, but anything that could reduce the workload would be worth considering
- support for reading/writing configurations to USB key. Installs can run unattended using configs stored on the USB key. This would allow you to install fleets of identical servers (e.g. Web farm) quickly
I'm sure there's other requirements you could come up with, but this would let you quickly put an entire data centre together. MS in particular would find it hard to compete with this.
Why didn't the company involved strip or mutilate all the sensitive data their contractors were given to work with? There's absolutely no excuse to give contractors sensitive data for this type of work.
Hell, even replacing every letter with a random letter and every number with a random number in sensitive fields would probably be enough; rather than "Mrs Joan Smith", they could've been working with "Grc Meas Fesze" without any impact. Change all names, addresses and "comments" and that would just about do it.
Doing this would completely remove the possibility of this problem happening.
If I was "Mrs Joan Smith", I'd be suing the retainer of my personal data on this basis. No wonder they're still considering whether to tell the victims what has happened...
I'm interested in getting involved against this becoming law, and I'll contact you offline. I'm particularly concerned about the stupidity of extending copyright to meet the so-called requirements of large US-based corporations.
m ason.html
However, it's important to realise that Australian common law offers built-in protection against a lot of the worst of the DMCA. Common law exists to remove the need for laws to deal with highly-specific information. It's one of the reasons why we don't get the continual barrage of new legislation that the US seems to get to vote on every few weeks.
Common law pretty much equates to "common sense". As an example, AFAIK there have been no prosecutions for illegal recording of TV shows - that's because we acquired certain common law rights when we purchased a VCR and one of those is the ability to use the "record" button.
Australian common law also protects (IIRC) e.g. decompiling code for the purposes of creating interoperability - it's no coincidence that projects such as Samba were started in Australia, as they are legally protected via a common law right to "use something you've purchased in a reasonable manner" (OK, that's oversimplifying, but it's not far from truth).
AFAIK, common law would "trump" any DMCA-like law that was introduced. Common law is enshrined in the Australian Constitution, and that isn't likely to be tweaked to accomodate Johnny-come-lately stuff like DMCAs.
Common law is one of the reasons why there's no overwhelming push for a US-style Bill Of Rights in Australia; many things that would be covered by it are already covered by common law. Any introduction of a Bill Of Rights here would probably only create grey areas that don't currently exist. Although there's very little to actually prevent it, we also don't have governments subjected to massive and well-organised financial lobbying from interest groups (e.g. large companies); such lobbying in the US is primarily driven by a proposed transfer of rights from individuals to those interest groups, and common law protects us as individuals to the point where these lobby groups have no reason to exist here.
Oh, IANAL but I've dated a few... A useful reference on common law is at http://www.austlii.edu.au/au/special/alta/alta95/
I'd use a carrot and stick approach.
Carrots:
- Offer better/more complete Thai language support in Win Lite.
- Offer it at low cost, as we now know will happen
Sticks:
- *Any* hardware upgrade results in needing to re-register with MS Thailand. More memory, a CPU upgrade; just keep re-registering every time. That way they can track usage that much better, and possibly get early warning of changes in usage trends.
- Possibly remove functionality to the point where Win Lite is essentially useless in corporate environments. Limit access to file servers, or network printers, or something similar; maybe time bomb it so you have X days to upgrade to the more expensive version once you start using "advanced" features
- Limit it to the point where it's essentially useless outside of Thailand. That shouldn't be hard given the uniqueness of Thai language
to find an anti-trust lawyer who hasn't worked for or against Microsoft these days, wouldn't it?
> One benefit to having email is the ability to post
> information anonymously in order to avoid possible
> repercussions.
The obvious, but possibly overly-simplistic, solution to this is to change the email programs we're using. Make them configurable to either of these settings:
- I'll only accept email with "warranties". This would be the "normal" setting
- I'll accept email from anyone
That would mean, if you send "anonymous email", there's no guarantee it's gonna be accepted at the other end. Hopefully, sufficient people would choose the first option to kill off the spam market.
Whenever an "I won't accept anonymous email" user connects to his/her mail server, the mail server could then ditch the unwanted email for that user.
As the parent points out, there are valid reasons to send anonymous email, but the recipients of those messages will normally know in advance that they'll be receiving anon email and can set themselves up to deal with it. I'm thinking crisis support groups, specific mailing lists and so forth.
*Maybe* this could work - I'm sure there's a few aspects that have escaped me at sparrow fart in the morning...
The ZDNet article points out that if all the "pirated" tracks in Australia were purchased for $A0.99, then the record companies would be $2b better off.
As of now, my understanding is that Apple sells tracks for $US0.99, and is in pretty close to a breakeven state for iTunes (this may have changed recently, as surely the sheer volume going through iTunes would cause them to move into profit at some point). Regardless, it seems that $US0.99 is pretty close to the breakeven point, and you'd assume the breakeven cost in Australia would be no lower than that given the population is so small - let's cut the record companies some slack and assume $US0.90 is the breakeven point for online music sales in Australia.
$A0.99 translates to $US0.76. Now, since it costs $US0.90 to provide a downloadable music track, Kazaa is actually *saving* the record companies $US0.14 per downloaded track. By my calculation, the 850,000 tracks downloaded via Kazaa haved saved Australian record companies $US119,000 in providing that service.
What's that? Bogus use of statistics, you say?...
> I met one once at an Open Source conference and ...So you're implying that the other people at the Open Source conference weren't total pariahs, then?
> when you ask him what he does he very plainly
> states "I'm a spammer". The guy was a total
> pariah.
Riiiight
Although I switched to Debian some time back, I still buy every 2nd or 3rd Mandrake release and keep it on the shelf. Of all the for-pay distributions, I like Mandrake's model the best and I think there'd be a big hole in the desktop market in particular if Mandrake was to disappear.
/home. I don't have to rebuild configs that were stored in registry keys on Window boxes and thus lost during a migration.
The reason I switched to Debian was Knoppix. The hardware detection is that good; the install time is tiny; the products on the CD are pretty close to the full set of what I want.
The reason I use Linux is the fact that, when I install it, it just works and keeps on working. My experience is that a Windows desktop has a working life of about 6 months; after that, you're faced with a box that runs slower and slower and more and more problems are occurring. The only way of fixing that is to reinstall it, which consumes about 2 days by the time you've installed every app, every service pack and found every serial number (note: we're talking home PCs here). I've got a Linux file server that's run essentially unchanged except for disc upgrades and security patches since 1998, and has never once given me a problem - that simply doesn't happen with Windows boxes.
Installing a Linux box is a breeze these days. From bare metal, I can get a Linux desktop box installed and fully working *with all applications* within about 1-2 hours. For servers, it's often less. If necessary, I can then kick off an apt-get upgrade, come back in an hour or so, reboot and the box is operational. From bare metal, unless you're installing from an image, it takes many more hours sitting in front of a Windows system to install apps and service packs. I hate that I have to keep clicking "OK" or typing in serial numbers, so I can't wander off and do something else while the box installs. If you've got e.g. Ghost, you're covered to some extent, although then you hit problems if your hardware config is significantly different from that used for the Ghost image.
I like the fact that my Linux boxes never have downtime due to viruses or other rubbish. Whether this is due to market share or not, I don't care - my desktop Linux box hasn't had downtime other than for hardware changes for years.
I like that I can pick up my personal config from a Linux box and migrate it to new hardware just by copying
...that the record companies might be reading this article and taking notes?
I wonder why Apple hasn't made more of this in their advertising. It seems that having possibly the highest-profile series of movies in many years put together using your gear would be worth telling people about.
If the article is accurate, it's a great example of working globally that a lot of Apple's potential customers might want to hear about.
It'd certainly attract more positive interest than those ridiculous "HP Invent" advertisements - they're just laughable. Every time I see a new one, I think "What the hell am I looking at?" which I suspect isn't the message HP wants to be getting across.
> What can you say or do in a forty minute long
> meeting?
Actually quite a lot. I'm assuming they didn't spend the first 20 minutes talking about the latest sporting results or TV programs.
I imagine the Pres could well have heard about a bunch of FOSS stuff, and, rather than reading through various interpretations of the GPL, thought it might be worth having a chat to the guy that wrote it. If I understand it correctly, the role of the President in India is largely to "suggest directions" for the country to go in, and I could well see this meeting as being pivotal in any additional embracing of FOSS in India.
I'd love to know the types of conversations Bill Gates has with world leaders, which seem to occur on a fairly regular basis. It's not as though MS actually employs loads of people worldwide, or directly contributes any sizeable chunk of loot to non-US economies, which is what's gonna be of interest to foreign heads of state. Sure MS supplies a lot of infrastructure stuff to governments, but you don't see e.g. the president of Ford Motor Co or Xerox getting meetings with world leaders on any regular basis. I just can't see that the conversation would be anything more involved than "buy my stuff" or "build this infrastructure so my stuff can do more things". Exactly why this type of conversation would merit the time of a world leader escapes me, yet it happens often enough.
> I've never seen a middle manager promoted from a
> technical job.
There was a time not so long ago when techos hit a form of glass ceiling in a company of any size. You could either sit at basically the same pay scale forever as a pure techo, slip sideways into consulting/marketing/sales or move into management. Strangely, the thinking was that the best/only person to manage techos was an ex-techo, so there was never any real shortage of those jobs around.
> The average middle manager wouldn't know a
> Postfix/procmail e-mail system if it crawled out
> of their ass, jumped up on their desk and did the
> tap number from 42nd street.
Agreed.
However, the middle manager who runs those systems know they're spending a sh1tload on trapping viruses and storing spam - or he would if the people working for him told him so.
He'll also know that ditching that spam and not getting nailed by viruses every few weeks will save him money.
I've got several mates who moved from geekdom into middle management over the years, and they're doing these sorts of changes. They might be far enough removed these days to not know about Postfix/procmail, but they're tuned in enough to see that it can solve their problems once the basic concepts have been explained to them.
They're seeing their teams increasing in size, while those of less technical managers are shrinking. They're also getting pay raises for themselves and their teams. Funny how solving business problems will do that for you...
> Corporate middle management is not interested in
> facts. They are not interested in improvement.
> They are not interested in efficiency that is not
> accomplished by either making people shovel shit
> or firing people.
> Middle management seeks to maintain the status
> quo, and to do nothing unless it is absolutely
> necessary. Incompetence, bankruptcy, waste,
> stupidity, anything is better than trying and
> failing.
I'll call bullshit on this.
Corporate middle management (MM) now faces the repercussions of CIOs and senior management telling shareholders "we'll reduce the cost of IT by 20/30/50% in the next 12 months". Middle management then gets told "do this or die"; either they slash their costs by A LOT within 12 months or they're out the door. Whatever was the case in the past, MM is now *all* about efficiency.
A sizeable chunk of MM has worked out "Hmm, if we keep doing what we have been doing, we'll keep getting the same results, so now we have to try something different". In many cases, they don't yet know what "something different" is or should be, but they are on the lookout for something - anything - that means they won't be leading their team into the unemployment office in 12 months' time.
**Now** is the best possible time to go to these MM guys with your ideas.
I'll give you an example: want to put e.g. Postfix/procmail in front of Internet-facing MS Exchange servers and use it to (a) de-evil incoming email with evil HTML content such as @ signs in URLs, and (b) filter out email from known open relays? Collect some figures on how much time/money has been lost in your org fighting spam and the latest HTML-based email virus, drop those figures on your MM's desk along with the costs of implementing your solution. If you do it right, your MM will realise, if it's done right, it'll slash his costs hugely and maybe get him a few percent closer to keeping his (and your) job intact.
The trick is to present data that makes sense to your MM. Don't tell him "we'll block 13432 incoming spam messages per day"; tell him "we'll block 13432 incoming spam messages per day that cost us $2300 per day in storage costs. My solution will cost us $3000 to implement, so it's paid for itself by the 2nd day". He has to talk in terms of financial outcomes, because that's what his boss wants; if you want to get your ideas across, you have to do the same.
Many techos, and I've done this myself in the past, present their ideas in such a way that it comes across as "It'd be really cool if we did X, and there might even be some benefits for the company if we did it. We're not quite sure exactly what X will cost, or how long it's gonna take to do it, but we should do it anyway because my geek buddy did it and he's really smart". It only took about 300 rejections before I worked out that this approach never works unless your boss has a goatee...
> Sure, most of us are pretty damn ugly but there's
> gotta be a few photogenic nerds among us. With
> good stories, too.
No, there aren't, at least not to Joe Average.
When one of us sleeps with J Lo, then we've got a good story
> Although it sounds like a racket, I have been told
> that professors don't get royalties for books sold
> at their own universities. This is to prevent the
> abuse that you just mentioned, which although
> cynical, is not true.
I'd love to see some proof of this.
When I did Comp Science at Melbourne University many years ago, it was remarkable how:
- for each class, there was exactly one recommended textbook
- the author of that textbook was the lecturer
- that textbook was only available at a single place, the MU Bookshop, and at a price that any sane person would call exorbitant even for an IT book
- new editions of those books appeared fairly regularly, with just enough changes so the page/chapter numbering was thrown out compared to the previous edition
- I'm fairly confident *none* of these textbooks were ever opened again once you'd finished that subject. They held very little of any use outside of passing that subject
> The United States Government has a duty to protect
> the interest of it's own citizens. It does *not*
> have the duty to ensure that Indians get wealthy.
The same argument works in reverse too. The Indian government has a duty to protect the interest of its citizens too, and I bet they'd be very interested in anything that could potentially derail their huge and growing IT businesses.
Consider: most US-based IT companies stopped being "US companies" a long time ago. Companies like Microsoft, Hewlett-Packard, etc are now global companies that started and are based in the US. They *don't* have to support their country of origin, or even their employees, as a "prime directive"; they're primarily answerable to their shareholders.
If Microsoft was to stop using Indian coders (disclaimer: I only assume they're doing so), then the Indian govt would go straight to MS India and tell them "give us back *our* jobs or we'll stop buying your product". MS India would then relay that to MS Redmond, who would then have to weigh up the potential loss in Indian sales vs. other alternatives. With the amount of dollars being talked about and the global nature of their customers and workforce, there's probably a good argument for MS to move out of the US - they could relocate their head office to some other country that didn't try to control where their workers came from.
Would the US government punish MS if they did this? Almost certainly, although that punishment would have to be based on not buying MS' products and the US govt is probably lagging behind govts elsewhere in the world in investigating the viability of non-MS software. Even though the US govt is the single biggest customer for MS, I'm not sure that the scope of punishment that would/could be dished out by the US govt would be sufficient to sway the decision.
If MS moved its base to e.g. India, what would the Indian govt do? Probably commit to buying more MS software, and probably allow MS to go into govt markets that they previously haven't had access to.
What would MS' current shareholders do? I bet they'd support it, on the basis that MS should be able to reduce their costs significantly and a loss of sales in the US should be compensated by additional sales in India (in the case where MS relocated to India).
I'm not saying that this is likely to happen, but I bet it's a scenario that the US govt and the large "US IT companies" have investigated fairly thoroughly.
> The biggest problems with knoppix are (1) it
> uses kde instead of gnome and (2) it has its own
> package structure that is incompatible with
> debian. So apt-get dist-upgrade or even apt-get
> upgrade will break everything. I've only had
> success upgrading individual packages with apt.
I've done both apt-get upgrade and apt-get dist-upgrade several times (over a period of a few months) on my installed-to-hard-disc Knoppix box, and haven't had a problem with it.
I've also installed an extra zillion games via apt-get for my kids to play on the same box, and they work fine too.
If you're having problems with this, it might be worth reporting it to the www.knoppix.net. The PC I used is a grey box clone running an old Celeron 533 with no "tricky" hardware whatsoever; maybe you're hitting problems with the specific hardware you're using.
SCO's documents continue to gloss over several key points:
- code that e.g. Andrew Tridgell writes for Samba cannot possibly be owned by SCO
- Samba code is released under the GPL. The GPL says, in effect, "I'm allowing you to use this code under a certain set of conditions". If the GPL is invalid, then basic copyright law will be in effect
- SCO distributes Samba code, presumably under the GPL. I'm guessing that SCO hasn't negotiated a separate agreement with the Samba guys to distribute their code under some non-GPL arrangement
- if SCO succeeds in getting the GPL rendered invalid (and that is unlikely since, in this example, it's the Samba guys saying "I'm gonna give you extra rights to use this provided you stick to these conditions..." which is a very common approach to licencing), then SCO is simply breaking basic copyright law in distributing Samba
In other words, if the GPL is valid, then SCO has no case. If the GPL is invalid, then SCO is breaking the copyright of lots of individual copyright holders. Either way, they lose
I remembered hearing Fels wasn't entitled to use the title "Professor" on the media, and Google turned up http://www.abc.net.au/mediawatch/transcripts/s8367 77.htm which, although inconclusive, implies that he *is* a Professor and is entitled to use the honorific. The confusion seems to have come about due to some out-of-date documentation supplied by a university in error.
D =514&SortType=6 seems to confirm it.
This: http://www-pso.adm.monash.edu.au/news/Story.asp?I
Apologies for listening to the media and believing what I hear. Anyone want to buy these SCO shares; I believe the price is headed for the moon!
It's a mystery how this "Professor" title got prepended to Fels' name. For all that he did a remarkable job while in charge of the ACCC, the "Professor" title is totally bogus - he can no more call himself "Professor" than an unqualified person can call their self "Doctor".
I don't need to track my kids using their phones - the RFID chips I implanted in their arms will do just fine, thanks.
Having worked on both sides of the fence, I can tell you: ..." conceivably might if you can give a good enough explanation
- IT managers value frankness. Yep, even the ones who give an unpleasant reaction to frank information appreciate the input, in my experience
- if you're gonna slam someone, you'd better be prepared to offer good justification for your opinions. "Fred is a dickhead" wouldn't get you a pleasant response; "Fred is a dickhead because
Norton Spam Filter 2004, now with haiku filtering! Guaranteed to filter 100% of spam, as long as the Internet doesn't resort to copyright infringement...
You know I really tried, but I just can't weave a SCO comment into this message...