I roughly estimate the ratio of SMS that actually reach people I sent (or vice versa) around 95% to 98%, once in a while one doesn't come through. However, during Xmas time or New Year,when many people seem to SMS a lot, or during big
events when people tend to SMS their beloved a lot, many SMS's get lost or get delayed, sometimes for even days!
OTOH, I saw virtually 90% loss of SMS with some providers when directly sending SMS using a modem from a computer accessing their SMS gateways with tools like smsclient rather then using a cell phone (not with all providers). This was a major annoyance when I used this feature as admin alert during network outage times... Either you got three clones of the same SMS, or you received none...
Ouch. I don't have SMS
Mmhhh.. Who owns a cell phone and does not SMS? And if you don't own one, what would one care anyway...?
(BTW, said things are based on SMS experience in Europe, Germany in particular...)
While usually almost every/.er is very fast at identifying corporate monopolistic evil (Microsoft, Intel, whatever) and pointing out how threatening they are, now many (are these the same people?) seem to be pissed off by the idea of "supporting" a company because it's their obligation to deliver a product which can compete both in quality and price with those who own the market (read Intel, plus AMD).What kind of support do they need? Are they having volunteers crusing through malls collecting spare change while singing Christmas songs, like the salvation army? Do they send you donation requests every three weeks with pityful photos of their starving employees? I don't think so, though some of the posts seem to imply this.
What they need is people and companies considering their fringy product despite the domination of that market by quasi-monopolists. Are there reasons to do so, apart from being an affectionate Linus-hugging idealistic person? Yes, there are!
The Transmeta chips, as said before, offer a much better performance/watt ratio than any Pentium or Athlon available, which is not only mandatory for portable/low-power devices, but also a plus in terms of <kyoto>sustainable use of energy resources.</kyoto> (Given, StrongArm and XScale play in that area, but they are not IA32-compatible..)
Transmeta so far will not implement the TCPA, AFAIK. If you dislike the TCPA (you dislike the TCPA, don't you?!), you need an alternative if you want to stay with the IA32-architecture (which will be around for some time despite upcoming IA64 and Hammer, IMHO).
I admit that power consumption and TCPA threats are (not yet) an issue for all of us who think electricity drips right out of the wall sockets, but if you ask yourself questions about energy and TCPA, TM can be part of the answer. The attitude of many posters which reads "Pah, I don't care, I can build a better system cheaper!" proves that the "support" they need is getting rid of preoccupation that the Intel/AMD/mobo companies cartel is the answer to all of our questions. The "support" they need is to realize that alternatives will vanish from a market which is dominated by few, very powerful companies unless you actively consider using the alternatives. There is good reason to consider TM and other alternatives, reason beyond some geekish affection and mere pity for a kewl silicon valley startup.
It's the same kind of "support" garage companies producing neat car mp3 players, *BSD delevopers who ask to occasionally buy a CD to support the costs, or the small grocery store next to your Walmart needs: If you care for the product, buy the product. If you don't care, then don't buy it, but don't be too surprised if it will be gone tomorrow due to a market which is occupied by few global players dominating their product area who do not need any customer support, because they a) have the monitary force to kick any small competitor out of the market, b) leave no option to the customers anyway, and c) have followers do not value alternative solutions for alternative questions.
Maybe we all can agree on this, because hardcore pro-capitalistic people usually love competition, and the rest of us love alternatives... >:)
-- "Where do you want to go today? Somewhere you can never take me." -- Chumbawamba
This isn't really the answer to your question, but since so many of us use these annoying small devices called cell phones, what about using your numerical remote control in the cell phone SMS fashion? (77726|666|6633777, | being a pause, for Ramones, for example) If you are ambitious and/or lazy, why not going for T9 support, having your MP3 player guess the choices you make...
May sound a bit uncomfortable, but its still smaller (and if you ask me, even geekier) than a full alphanumeric keyboard thing...
Don't get me wrong, I rather do not advocate PPTP as the best solution for secure wireless networking. I would opt for something which has proven (or belived) to be more secure, like IPSec.
However, I know people deploy PPTP-based VPN networks, for several reasons, among them PPTP being a free (as in beer) VPN version for Windows versions since Windows 95. I know of a university WLAN network which has to deal with numerous clients the network admins don't control and thus have to support as many client versions as possible. They chose to use longer passwords, which function as 'keys' in MS PPTP.
It's true that the challange/response authentication scheme has been proven to be pretty insecure, mainly because LAN Manager passwords are suffering from significantly weaker encryption than NT passwords, but for backward compatibility reasons both are sent together, always, which makes password guessing way more easy. I figure that is what you refer to as 'breaking keys'. I agree. OTOH, Microsoft released a so-called pptp3-fix, which fixes the LAN Manager password problem. I know there remain a number of other issues, but the main problem has been fixed in there, according to MS and some other sources.
Still, I am no fan of PPTP. But sometimes, admins face needs which force them to make compromises...
Before deploying PPTP, you might want to have a look at these security issues. Others consider it secure as long as the keys are sophisticated enough, IIRC...
Probably you will not expect peak performance from that anyway... So here is what I would go for:
Linux Kernel 2.2 (with low memory i would recommend 2.2..) or a really stripped-down 2.4 kernel, running a modern distro which is rather slick in itself, like Slackware or Debian. This will help you avoiding numerous security holes in older distros.
There has been an article which focusses on small yet functional destop programs.
If you want to go with really stripped-down distros, which are suitable (or optimized) for embedded computers, check this link.
I have to agree with some of the other posters that oneofthe *BSD derivates can be and feel a lot smaller than full-featured, KDE3-based Linux distros...
If security is not much of an issue for you, for whatever reason, you might want to go for an outdated Linux distro. Watch out for a 2.0 or 2.2 kernel, and libc5 instead of glibc2/libc6, or you might not gain much from the old stuff... Or even Minix?
VSTa?
I have been running couple of very humble 486-based boxen with some of these things lately for quite some time as well, it can be a nice and productive experience if you adjust to the capabilities of the granny hardware.. And it surely teaches you patience.)
... for nerds or stuff that matters? I haven't watched it, among the reasons for ignoring it was the fact that this must be the fourth or fifth of these domino shows, one more boring than the other, and IIRC all of them conducted by the Endemol TV company which brought us other fine and amazing stuff (note the irony, please) like Big Brother...
What's next? A/. post on the Biggest Card House In The World Live TV show? A story on My Cat Can Jump Wider Than Yours afternoon tv feature? Or what about these shows where neurotic parents make their kids dress up like Ms. Spears or Mr. Jackson and mimic to their songs? Maybe we need a new Section here, called Irrelevant TV Shows which I can ignore...
(In fact, I might want that Mini Playback Show stuff to be mentioned here as soon as some geek parents make their son dress up like RMS and have him sing this song... That would be worth a video tape. But that's a different thing, right?.)
They even state that it isnt a replacement for 802.11.
I agree , just too slow for that. But I can imagine many useful with services with low bandwidth, esp. if these slow devices could mesh together to some kind of community network: grassroots SMS, mostly email is fine on slow connections, and even browsing the web with something like lynx or w3m would be doable. But for that these things are way too expensive... Maybe I should reconsider going on with my CB/walkie-talkie network idea...
I have been paranoid about moving away from ext2 in production environment myself, so I have been trying out ext3 and reiserfs on personal boxes for the last eigth months. No real problem so far (once I still had a fs error on reiserfs after replaying the log, a fsck fixed that). These are no hard hit or database servers, but I tend to copy a lot of files and data (several gigs) simultaneously between reiserfs file systems and mix up my cd-rom eject button with the reset button from time to time, and on the ext3 box I tried hacking some usb driver which made it crash every ten minutes over three weeks...
Being concerned for reliability, you might look into the ext3 options to have ordered writes etc., or otherwise your meta data might be fine, although your actual file data might get screwed... The kernel mailing list should hold hints on stability (the more absent complaints are, the more stable it is;), and as said before, in the end you have your backups... Don't you?
I still wonder about meta journaling on database servers with huge data files, and the next thing I am paranoid about is using LVM on Linux:)
I bet this has been said here before, but generally increasing security often includes breaking applications by definition. Like application proxies and firewalls, which purposely break some network funtionality in order to secure the network. And for instance, removing the double-click-and-Word-will-open feature for Microsoft Word documents in Microsoft Outlook, which has caused lots of havoc, will break a major convenient functionality of Outlook, from the view
of its users.
So, <paranoid disclaimer>whatever Microsoft is implying when they say that they will break applications</paraoind disclaimer>, it is always "Give me convenience or give me security" (Kudos to these fine guys), otherwise we wouldn't use passwords, encrypted authentication and other inconvenient stuff etc. "Why not just skip all these logins? They make my brain hurt from all the stuff I need to remember..."
So again, either you demand more knowledge, responsibility and work from the user, or you leave all the necessary security decisions to the software... There is a lot of reason for criticizing Microsoft in many ways, but I think its quite unrealistic to ask for ultimately convenient, ultimately secure software simultaneously... Consequently, either bash them for being insecure or for giving up convenience, please don't do both at the same time, because that doesn't seem to make much Sense(TM) to me....)
But still, it is a wonderful tool that applies in the current situation:)
It's awesome, isn't it? But last time I checked apt-get didn't support automatic cryptographic signature checks on install... I don't mean md5sums, because unsigned hashes are as easily spoofable as the data itself, I mean PGP/GnuPGP-like, PKI-authorized sigs. Maybe I just don't know of that apt-get feature, personally I cannot believe such an advanced tool like apt-get lacks such a vital, security-related feature... IIRC, rpm does have such a thing...,)
Oh, well, funny.... Then I might have missed something... Maybe I should have done some apt-get voodoo on my BrainOS before reading... )
apt-get is nice, but its not the solution to every problem, especially not the fundamental problem of network and computer security, not the roots solution for trojaned security software, and especially not the solution for people starving in so-called not-so-developed countries, the greenhouse effect (you remember, the european conspiracy to bring down US economy?) or nuclear waste... Although members of the Church of Apt-Getarism and the Gentoo Temple always point out that their l33t distro-specific rocket science saves us from all evil...
Good your apache is apt-secure now... You are sure you had installed that package before?
And for the record: Now I was being funny, wasn't I;)
apt-get update...
well, I have not installed these sniffing proggies, so it should be okay.
Darn... apt-get even makes your box more secure than before even if you haven't actually installed the bad packages? This must be the Holy Grail! And it should be okay? Not only that you have not installed tcpdump and libpcap, what definitely makes it okay, you don't even trust apt-get to really solve your (non-existing) problem... Now I wanna join the apt-get cult... Where can I register?
I bet you recommend penicillin over other medicine even when you got no infection! Or do you use apt-get then as well? Doesn't make any difference anyway...
(For the record: I use Debian GNU/Linux among other stuff...)
Like three years ago I came across a LyX (the WYSIWYG-frontend for TeX) developer at a LUG who claimed that somebody of them is developing a vt100-based terminal version of LyX... I am not sure if this yielded any results, but maybe you can search the LyX lists for that or contact the developers...
Probably a worthless because outdated info that I am giving, but such a thing would really be interesting I think, and it's a nice deigital urban legend anyway...:)
Well, from what I read in the original question users now are solely authenticated by their MAC addresses. Myself, I could cope with entering passwords into web forms (or I would find a way to automate that), I just think that there would be users that might not at all find web forms handy, despite the ability to have passwords stored by browsers... My main point was to point to StockholmOpen anyway...;)
Sounds pretty much like what these guys do... They offer their implementation for download as well. Pretty cool stuff they do, IMHO. I am just not sure if this rather manual authentication method (entering account name and password into some login form) will be accepted well by customers... The StockholmOpen people also note that their system is not 100% secure, but its way better than doing authentication on the basis of MAC addresses.
Quick File Distribution Challenge on Advogato
on
Ghost for Unix
·
· Score: 2, Interesting
There has been this post at advogato couple of weeks ago which is about distributing huge amounts of data to many machines within a hilarious time... Though the solution implied by the author has not been revealed on there, it's quite an interesting read. The challenger excludes multicasting in order to make his question harder, but some posters there refer to multicast anyway...
Personally, I agree with UDP multicasting being the way for multiple network-based clones... For only a handful of clones Mondo+Mindi might be an alternative, too... No network, but CD-ROMs over sneakernet though...:)
Mail size limits everywhere!
on
E-Mail Size Limits?
·
· Score: 2, Interesting
As many other postings suggested, 2 megs is a widely deployed limit. Whenever you think about your own corperate mail size limit, keep in mind that even a 100 meg limit won't help you if the sites you send mail to have a way smaller limit. Since you never know what limit your mails will hit on the recipients' (or forwarding) servers, always keep it considerably small, or it will be rejected by the recipent's server... Better have it being rejected within your own network, so you don't waste precious bandwidth to the outer world...
I saw people sending out their whole mp3 collections in 30 meg-sized mails, I saw others suffer from 5 megs holiday pics in their mailboxes accessed by humble 56k modems and having to give them a POP3 primer on the phone to have them delete that stuff... Yes, mail isn't ment for big files, and besides this orthodox opinion, it still isn't very practical to send huge files per mail.
I would go for the send-links-in-mails-to-files-on-{web,file}servers idea, maybe protected through SSL, temporary passwords and having the files auto-deleted after three days or when having been fullydownloaded one or two times or whatever... (Proposing some minor access security for the files being made available, I wonder how many of these suits 'needing' unlimited mail sizes protect their files when sent by mail? I have never seen a single member of this demandish breed (being subject to this publication) being able to handle PGP/GnuPG/whatever, they often consider that as another evil trick by notorious sysadmins, since security measures hinder them in doing their work properly, just as mail size limits do... So much for ranting:)
First they invented the CPU-ID when producing the Pentium, but I didn't care since my BIOS gave me the option to disable it.
Then they put some GUID in MS Word documents, but I didn't care since I used StarOffice/LaTeX/Applix.
Then they would require a personal registration for any XP install, but I have been running Linux/*BSD so I didn't care.
Then they put "DRM" into Media Player 9, but since I ran mplayer and xmms, I didn't care
Then they started putting obscure crypto hardware comtrolled by some vendors' consortium into each and every PC and I could not run my free software, but there was nobody to help me, since my emails and postings were denied by the TCPA platforms to their owners, and nobody could read then anymore...
<disclaimer>Rev. Niemöller was facing worse conditions when writing the original, I know that...</disclaimer>
Rate me off-topic, but (esp. considering MIT's OpenCourseWare) I think I would love to attend that meeting virtually or watch some digitally available copy of that session... Is that possible?:)
And if I could pose a question, I would probably ask how they would try to fight the immanent problem that people always find a way to beat copy protection, since the beginning of sold 8" floppy disks, and there will be no way to prevent that in Palladium either, I swear...
There, see? not only is everything set, we even have two complete different sets of answers
If I were a moderator, I would mod him up, since this is one of the major problems for non-geek users: consistency. There is freedom in multiple, rather incompatible or at least inconsistent alternatives, but there is also complexity. Myself, I like having the choice and do not want to trade it for simplicity, but then I have to accept Linux will never be a system for people who just want to use it.
Yeah, great idea, its up to you to start something similar in your local area, to support the kids in YOUR country now.
And BTW, this is not america, but I have seen schools which got computer stuff donated being *very* picky about what they want, pentium 100 with 1 gig drives would be at the low end of their expectations, if they would take it at all... It does not run Windows XP and/or multimedia stuff properly, you know...
Slashdot Mirror
I roughly estimate the ratio of SMS that actually reach people I sent (or vice versa) around 95% to 98%, once in a while one doesn't come through. However, during Xmas time or New Year,when many people seem to SMS a lot, or during big events when people tend to SMS their beloved a lot, many SMS's get lost or get delayed, sometimes for even days!
OTOH, I saw virtually 90% loss of SMS with some providers when directly sending SMS using a modem from a computer accessing their SMS gateways with tools like smsclient rather then using a cell phone (not with all providers). This was a major annoyance when I used this feature as admin alert during network outage times... Either you got three clones of the same SMS, or you received none...Ouch. I don't have SMS
Mmhhh.. Who owns a cell phone and does not SMS? And if you don't own one, what would one care anyway...?
(BTW, said things are based on SMS experience in Europe, Germany in particular...)
... they need your support..
While usually almost everyWhat they need is people and companies considering their fringy product despite the domination of that market by quasi-monopolists. Are there reasons to do so, apart from being an affectionate Linus-hugging idealistic person? Yes, there are!
- The Transmeta chips, as said before, offer a much better performance/watt ratio than any Pentium or Athlon available, which is not only mandatory for portable/low-power devices, but also a plus in terms of <kyoto>sustainable use of energy resources.</kyoto> (Given, StrongArm and XScale play in that area, but they are not IA32-compatible..)
- Transmeta so far will not implement the TCPA, AFAIK. If you dislike the TCPA (you dislike the TCPA, don't you?!), you need an alternative if you want to stay with the IA32-architecture (which will be around for some time despite upcoming IA64 and Hammer, IMHO).
I admit that power consumption and TCPA threats are (not yet) an issue for all of us who think electricity drips right out of the wall sockets, but if you ask yourself questions about energy and TCPA, TM can be part of the answer. The attitude of many posters which reads "Pah, I don't care, I can build a better system cheaper!" proves that the "support" they need is getting rid of preoccupation that the Intel/AMD/mobo companies cartel is the answer to all of our questions. The "support" they need is to realize that alternatives will vanish from a market which is dominated by few, very powerful companies unless you actively consider using the alternatives. There is good reason to consider TM and other alternatives, reason beyond some geekish affection and mere pity for a kewl silicon valley startup.It's the same kind of "support" garage companies producing neat car mp3 players, *BSD delevopers who ask to occasionally buy a CD to support the costs, or the small grocery store next to your Walmart needs: If you care for the product, buy the product. If you don't care, then don't buy it, but don't be too surprised if it will be gone tomorrow due to a market which is occupied by few global players dominating their product area who do not need any customer support, because they a) have the monitary force to kick any small competitor out of the market, b) leave no option to the customers anyway, and c) have followers do not value alternative solutions for alternative questions.
Maybe we all can agree on this, because hardcore pro-capitalistic people usually love competition, and the rest of us love alternatives... >:)
--
"Where do you want to go today? Somewhere you can never take me." -- Chumbawamba
Becouse thoose small annoying devices don't have number pads with letters overloaded?
Well, you can rave a lot of harrle when you put rome characterr on the rame porition. Like R and R, for example...This isn't really the answer to your question, but since so many of us use these annoying small devices called cell phones, what about using your numerical remote control in the cell phone SMS fashion? (77726|666|6633777, | being a pause, for Ramones, for example) If you are ambitious and/or lazy, why not going for T9 support, having your MP3 player guess the choices you make...
May sound a bit uncomfortable, but its still smaller (and if you ask me, even geekier) than a full alphanumeric keyboard thing...
Don't get me wrong, I rather do not advocate PPTP as the best solution for secure wireless networking. I would opt for something which has proven (or belived) to be more secure, like IPSec.
However, I know people deploy PPTP-based VPN networks, for several reasons, among them PPTP being a free (as in beer) VPN version for Windows versions since Windows 95. I know of a university WLAN network which has to deal with numerous clients the network admins don't control and thus have to support as many client versions as possible. They chose to use longer passwords, which function as 'keys' in MS PPTP.It's true that the challange/response authentication scheme has been proven to be pretty insecure, mainly because LAN Manager passwords are suffering from significantly weaker encryption than NT passwords, but for backward compatibility reasons both are sent together, always, which makes password guessing way more easy. I figure that is what you refer to as 'breaking keys'. I agree. OTOH, Microsoft released a so-called pptp3-fix, which fixes the LAN Manager password problem. I know there remain a number of other issues, but the main problem has been fixed in there, according to MS and some other sources.
Still, I am no fan of PPTP. But sometimes, admins face needs which force them to make compromises...
Before deploying PPTP, you might want to have a look at these security issues. Others consider it secure as long as the keys are sophisticated enough, IIRC...
(Yes, I know, those were ants... Anyway... .)
- Linux Kernel 2.2 (with low memory i would recommend 2.2..) or a really stripped-down 2.4 kernel, running a modern distro which is rather slick in itself, like Slackware or Debian. This will help you avoiding numerous security holes in older distros.
- There has been an article which focusses on small yet functional destop programs.
- If you want to go with really stripped-down distros, which are suitable (or optimized) for embedded computers, check this link.
- I have to agree with some of the other posters that one of the *BSD derivates can be and feel a lot smaller than full-featured, KDE3-based Linux distros...
- If security is not much of an issue for you, for whatever reason, you might want to go for an outdated Linux distro. Watch out for a 2.0 or 2.2 kernel, and libc5 instead of glibc2/libc6, or you might not gain much from the old stuff... Or even Minix?
VSTa?
I have been running couple of very humble 486-based boxen with some of these things lately for quite some time as well, it can be a nice and productive experience if you adjust to the capabilities of the granny hardware.. And it surely teaches you patienceWhat's next? A /. post on the Biggest Card House In The World Live TV show? A story on My Cat Can Jump Wider Than Yours afternoon tv feature? Or what about these shows where neurotic parents make their kids dress up like Ms. Spears or Mr. Jackson and mimic to their songs? Maybe we need a new Section here, called Irrelevant TV Shows which I can ignore...
(In fact, I might want that Mini Playback Show stuff to be mentioned here as soon as some geek parents make their son dress up like RMS and have him sing this song... That would be worth a video tape. But that's a different thing, right? .)
I agree , just too slow for that. But I can imagine many useful with services with low bandwidth, esp. if these slow devices could mesh together to some kind of community network: grassroots SMS, mostly email is fine on slow connections, and even browsing the web with something like lynx or w3m would be doable. But for that these things are way too expensive... Maybe I should reconsider going on with my CB/walkie-talkie network idea...
I have been paranoid about moving away from ext2 in production environment myself, so I have been trying out ext3 and reiserfs on personal boxes for the last eigth months. No real problem so far (once I still had a fs error on reiserfs after replaying the log, a fsck fixed that). These are no hard hit or database servers, but I tend to copy a lot of files and data (several gigs) simultaneously between reiserfs file systems and mix up my cd-rom eject button with the reset button from time to time, and on the ext3 box I tried hacking some usb driver which made it crash every ten minutes over three weeks...
Being concerned for reliability, you might look into the ext3 options to have ordered writes etc., or otherwise your meta data might be fine, although your actual file data might get screwed... The kernel mailing list should hold hints on stability (the more absent complaints are, the more stable it is ;), and as said before, in the end you have your backups... Don't you?
I still wonder about meta journaling on database servers with huge data files, and the next thing I am paranoid about is using LVM on Linux :)
So, <paranoid disclaimer>whatever Microsoft is implying when they say that they will break applications</paraoind disclaimer>, it is always "Give me convenience or give me security" (Kudos to these fine guys), otherwise we wouldn't use passwords, encrypted authentication and other inconvenient stuff etc. "Why not just skip all these logins? They make my brain hurt from all the stuff I need to remember..."
So again, either you demand more knowledge, responsibility and work from the user, or you leave all the necessary security decisions to the software... There is a lot of reason for criticizing Microsoft in many ways, but I think its quite unrealistic to ask for ultimately convenient, ultimately secure software simultaneously... Consequently, either bash them for being insecure or for giving up convenience, please don't do both at the same time, because that doesn't seem to make much Sense(TM) to me... .)
Neither did I... .)
But still, it is a wonderful tool that applies in the current situation :)
It's awesome, isn't it? But last time I checked apt-get didn't support automatic cryptographic signature checks on install... I don't mean md5sums, because unsigned hashes are as easily spoofable as the data itself, I mean PGP/GnuPGP-like, PKI-authorized sigs. Maybe I just don't know of that apt-get feature, personally I cannot believe such an advanced tool like apt-get lacks such a vital, security-related feature... IIRC, rpm does have such a thing... ,)
apt-get is nice, but its not the solution to every problem, especially not the fundamental problem of network and computer security, not the roots solution for trojaned security software, and especially not the solution for people starving in so-called not-so-developed countries, the greenhouse effect (you remember, the european conspiracy to bring down US economy?) or nuclear waste... Although members of the Church of Apt-Getarism and the Gentoo Temple always point out that their l33t distro-specific rocket science saves us from all evil...
Good your apache is apt-secure now... You are sure you had installed that package before?
And for the record: Now I was being funny, wasn't I ;)
well, I have not installed these sniffing proggies, so it should be okay.
Darn... apt-get even makes your box more secure than before even if you haven't actually installed the bad packages? This must be the Holy Grail! And it should be okay? Not only that you have not installed tcpdump and libpcap, what definitely makes it okay, you don't even trust apt-get to really solve your (non-existing) problem... Now I wanna join the apt-get cult... Where can I register?
I bet you recommend penicillin over other medicine even when you got no infection! Or do you use apt-get then as well? Doesn't make any difference anyway...
(For the record: I use Debian GNU/Linux among other stuff...)
Like three years ago I came across a LyX (the WYSIWYG-frontend for TeX) developer at a LUG who claimed that somebody of them is developing a vt100-based terminal version of LyX... I am not sure if this yielded any results, but maybe you can search the LyX lists for that or contact the developers...
Probably a worthless because outdated info that I am giving, but such a thing would really be interesting I think, and it's a nice deigital urban legend anyway... :)
Well, from what I read in the original question users now are solely authenticated by their MAC addresses. Myself, I could cope with entering passwords into web forms (or I would find a way to automate that), I just think that there would be users that might not at all find web forms handy, despite the ability to have passwords stored by browsers... My main point was to point to StockholmOpen anyway... ;)
Sounds pretty much like what these guys do... They offer their implementation for download as well. Pretty cool stuff they do, IMHO. I am just not sure if this rather manual authentication method (entering account name and password into some login form) will be accepted well by customers... The StockholmOpen people also note that their system is not 100% secure, but its way better than doing authentication on the basis of MAC addresses.
Personally, I agree with UDP multicasting being the way for multiple network-based clones... For only a handful of clones Mondo+Mindi might be an alternative, too... No network, but CD-ROMs over sneakernet though... :)
I saw people sending out their whole mp3 collections in 30 meg-sized mails, I saw others suffer from 5 megs holiday pics in their mailboxes accessed by humble 56k modems and having to give them a POP3 primer on the phone to have them delete that stuff ... Yes, mail isn't ment for big files, and besides this orthodox opinion, it still isn't very practical to send huge files per mail.
I would go for the send-links-in-mails-to-files-on-{web,file}servers idea, maybe protected through SSL, temporary passwords and having the files auto-deleted after three days or when having been fullydownloaded one or two times or whatever... (Proposing some minor access security for the files being made available, I wonder how many of these suits 'needing' unlimited mail sizes protect their files when sent by mail? I have never seen a single member of this demandish breed (being subject to this publication) being able to handle PGP/GnuPG/whatever, they often consider that as another evil trick by notorious sysadmins, since security measures hinder them in doing their work properly, just as mail size limits do... So much for ranting :)
First they invented the CPU-ID when producing the Pentium, but I didn't care since my BIOS gave me the option to disable it.
Then they put some GUID in MS Word documents, but I didn't care since I used StarOffice/LaTeX/Applix.
Then they would require a personal registration for any XP install, but I have been running Linux/*BSD so I didn't care.
Then they put "DRM" into Media Player 9, but since I ran mplayer and xmms, I didn't care
Then they started putting obscure crypto hardware comtrolled by some vendors' consortium into each and every PC and I could not run my free software, but there was nobody to help me, since my emails and postings were denied by the TCPA platforms to their owners, and nobody could read then anymore...
<disclaimer>Rev. Niemöller was facing worse conditions when writing the original, I know that...</disclaimer>
Rate me off-topic, but (esp. considering MIT's OpenCourseWare) I think I would love to attend that meeting virtually or watch some digitally available copy of that session... Is that possible? :)
And if I could pose a question, I would probably ask how they would try to fight the immanent problem that people always find a way to beat copy protection, since the beginning of sold 8" floppy disks, and there will be no way to prevent that in Palladium either, I swear...If I were a moderator, I would mod him up, since this is one of the major problems for non-geek users: consistency. There is freedom in multiple, rather incompatible or at least inconsistent alternatives, but there is also complexity. Myself, I like having the choice and do not want to trade it for simplicity, but then I have to accept Linux will never be a system for people who just want to use it.
It must be 2.6, versions ending in a zero are always less stable than those having higher minor version numbers and thus less trustworthy... :)
And BTW, this is not america, but I have seen schools which got computer stuff donated being *very* picky about what they want, pentium 100 with 1 gig drives would be at the low end of their expectations, if they would take it at all... It does not run Windows XP and/or multimedia stuff properly, you know...