Sections: Main page Linux in the news Security Kernel Distributions Development Commerce Announcements Back page All in one big page See also: last week's Back page page. |
Linux links of the weekSome say that Linus is God; others may well be more interested in the Bible for Linux Project. They are working to promote the development of Bible software for our favorite operating system. The Review of Operating Systems is a simple and lengthy page which attempts to present a comprehensive list of operating system resources on the net. There's a lot of stuff out there... Section Editor: Jon Corbet |
August 26, 1999 |
|
Letters to the editorLetters to the editor should be sent to letters@lwn.net. Preference will be given to letters which are short, to the point, and well written. If you want your email address "anti-spammed" in some way please be sure to let us know. We do not have a policy against anonymous letters, but we will be reluctant to include them. | |
Date: Thu, 19 Aug 1999 15:35:02 +0200 From: Bernd Paysan <bernd.paysan@gmx.de> To: letters@lwn.net Subject: RHAT overpriced? Well, aside from the fundamentalist opinion, that all internet stocks, and certainly Red Hat are way overpriced, there's some logic behind it. After all, Amazon or Yahoo aren't priced that high for their business they make now, or for the class of business they do (everybody can build a web-based bookstore or a portal site, lots do), but for building up a brand name. $5 billion for a brand name isn't overpriced (supposed that this thing becomes a commodity), perhaps $50 billions are. Look at traditional brand names like Levis. Everybody can make jeans, after all, it's nothing more than blue cotton and yellow thread; jeans are really a generic product. In fact, a lot of companies do make jeans, yet brand name jeans can sell at a premium price, although the only difference (to quality generic jeans) is the label. The same goes for Linux. Everybody can make a distribution, even a close rip-off or immitation of a brand name Linux (e.g. Mandrake is mostly Red Hat). The idea then is that even though there are a lot of "cheap immitations", the masses still will choose the brand. The main problem that contradicts that logic is that these things change in "Internet time". A few years ago, Slackware was the dominant distribution. Next year, Corel could be the dominant distribution. Or SuSE, when they continue to expand in the USA; they already are dominant in Europe. I don't think any of those can maintain a first class brand name for as long as Levis did. After all, none of them invented Linux, so nobody could write "The original Linux" on their package. And the value of the brand really depends on becoming a commodity (i.e. "Total world domination(tm)"), or at least a luxury good (such as Rolls Royce or Lamborghini). -- Bernd Paysan http://www.jwdt.com/~paysan/ | ||
Date: Thu, 19 Aug 1999 10:04:26 -0400 (EDT) From: "Timothy D.J. Hunt" <tim@NewExpression.com> To: letters@lwn.net Subject: Linux IPO's and the Linux Community Sir, I am been watching with some frustration as Redhat attempted to do the right thing by the Linux community, and ended up being frustrated by market regulations intended to protect the naive investor. Given that we are expecting many more successful Linux based IPO's, and given that each Linux based company should be attempting to return some of the wealth to the community that created it, how could we prevent this happening again? I have a couple of suggestions: 1) How about setting up a Linux foundation that Linux based companies can make stock grants to, pre IPO. The foundation can then funnel the wealth to the Linux community by sponsoring new development, and/or grant awards to significant contributors (similar to the MacArthur foundation). This also has the potential to create a virtuous cycle, since any new development that adds value to Linux, also adds value to Linux based companies. 2) Maybe an enterprising mutual fund company could setup a Linux based mutual fund with membership offered to significant contributors within the Linux community. Linux companies can then allocate some of their community stock to this mutual fund, without have to deal with the issues involved in offering stock to lots of individuals. Regards Tim Hunt timh@usa.net | ||
From: "Jim Watts" <jamescwatts@hotmail.com> To: lwn@lwn.net Subject: Best and worst CD from the Linux World show. Date: Thu, 19 Aug 1999 19:29:49 PDT Hey guys, I just finished going thru your Linux Show review, and while I was disappointed not to see a picture of myself (in my very rare Amiga shirt), I was even more disappointed to see you choose Amdahl's CD as the worst of the show... and then drag them thru the mud for it. To be honest I haven't even looked at my copy yet. But I have looked at Storm, Turbo, and SuSE, and none of those would install on my laptop, only Red Hat did. So I have no use for any of them either. I not going to try to excuse the copyright bull on the back. It was obviously a canned blurb they got from legal... and we all know what we should do with them. You do have to give them some credit for even trying. Fujitsu is not a small company (Amdahl's owners) and for Amdahl and Fujitsu to support Linux can not be a small thing for them... or the Linux community. Maybe the best way to get them to realize the error was to post a flaming review like yours, maybe not. At least they are trying, you have to give them that. If they can get Linux into just one shop where it has never been before, they've done well enough in my book. And if 10 more companies do the same, so much better. Hopefully the next public display from Amdahl will be better. Time tells all. How would you like to see Linux running on a 2000Mips (a Millenium) mainframe? That would be a nice demo. Cluster THAT! Just my 2 cents. (Have you noticed that the cent key is gone? Where'd it go?) -Jim | ||
Date: Thu, 19 Aug 1999 12:04:56 +0100 From: Sid Boyce <szb50@amdahl.com> To: letters@lwn.net Subject: RE: The CD Giveaways Amdahl Corporation is still around with excellent products to sell. Without fanfare, quite a number of us have been avid Linux users going way back, many machines dotted around the Corporation are running Linux for serious work. On a visit to our Sunnyvale HQ in January, I was using a Linux machine for 3 days before I knew it was Linux, I had never seen Window Maker before, that's one of the beauties of this great Linux we know and love. This was a hastily put together CD and from my soundings, the standard licensing document should definitely not have been used - the equivalent of a kernel oops perhaps ?. Rest assured, we are serious, we are keen to get going, we feel the buzz. We've got some serious and solid hardware in TeamServer and Elvis (RAID) and we shall have Linux solutions in the Corporate sector to complement our excellent Mainframe offerings. Most people at Amdahl are still new to Linux, but the pin is dropping rapidly, for me, I've been with Linux since it first appeared and have preached it to all. At first I gained the impression they thought I was a total eccentric, now I am something of a prophet perhaps. Everyone knows the sort of reaction you got on mentioning Linux, one colleague installed it on a Quad P-II which we sold with NT, when the Dept. manager passed by, he said "gosh, that's quick", "what's that ?", having been told it was Linux, he put his nose in the air and marched of saying, "that's not one of our supported operating systems". We get the last laugh!. Regards -- ... Sid Boyce...Amdahl(Europe)...44-121 422 0375 Any opinions expressed above are mine and do not necessarily represent the opinions or policies of Amdahl Corporation. | ||
Date: Fri, 20 Aug 1999 11:38:01 -0700 From: Anand Srivastva <anand@nmi.stpn.soft.net> To: webmaster@osopinion.com, letters@lwn.net Subject: Linux won't dominate the desktop Hi, I don't agree with you that Linux won't succeed on the desktop. Its like saying MSDOS can't work on the Desktop, remember how bad MSDOS was initially. The only thing wrong with Linux is that its too new. How much time has it been since people started even thinking of a GUI for Linux, just two years. Remember how much time MSDOS took to have a GUI. How long MS had to work to get a decent Windows GUI. How long it had failed, before it started succeeding. Just wait for a couple of years and Linux will be better. People don't want to blame someone. They want to get their work done. If there are problems then they want to be able to call up people who can solve the thinks. I think as Linux gets more mainstream people will find someone around them who knows about it. There are already companies who can help them with their questions. At the end its been asked a lot, but have you ever called up MS support. I have seen a lot of support for Windows, and the only solution that there ever is reboot or reinstall. I think people will appreciate if the problem can be really cured. Also the freedom they get when they can choose their support company. If Compaq doesn't do a good job, call up IBM ;-). I think, it is on the desktop that monetory price becomes very crucial. For the Corporate desktop you maybe right but for the common user, monetory price is very important. And when we generally talk about desktop users we are mostly talking about the common user. For them anything, linux's cost effectiveness will matter a lot. When the price of the PC with Linux is $400 and with Windows you have to add another $100, I think that $100 will mean a lot. Of course this includes the disclaimer that Linux is not yet very feasible for the common user. But the time is not very far away, maybe another year maybe some more, when Linux will become very useable by the common user. BTW, there is a notepad app for linux, and MS notepad I think works through WINE on linux. Linux just needs time, -anand | ||
Date: Mon, 23 Aug 1999 19:29:53 -0700 From: John Gardiner Myers <jgmyers@netscape.com> To: letters@lwn.net Subject: ESR's 'Will You Be Cracked Next?' I am astounded by claims that Linux is "invulnerable to macro attacks" and "immune to viruses." They are simply false. GNU Emacs will, upon loading text files, execute elisp code embedded in certain ways; it is vulnerable to macro attacks. Unix viruses have existed since 1989, an excellent reference is http://www.cyber.com/papers/plausibility.html | ||
To: Matt.Wilkie@gov.yk.ca, letters@lwn.net Subject: re: Back Orifice Date: Mon, 23 Aug 1999 02:31:13 -0400 From: Piotr Mitros <pmitros@MIT.EDU> > Although Linux can laugh at Back Orifice itself, Linux is -not- > immune to a BO style of attack: a trojan wrapped inside an innocuous > program which the recipient/user does want to run. (rootkits > anyone?) Despite Eric's flair for unsubstantiated hype, he is correct here. The major hole demonstrated by Back Orifice is not merely the idea of a trojan horse. The real problem it is meant to show is that if any (non-administrative) account is compromised, the whole system is compromised. This means that if a secretary in some company is fooled into running a trojan program, the accounts of all future users to log in on that machine can automatically be compromised. By sticking a small program in their startup files, any other machines they log in on can be compromised. This can easily spread over a corporate network like wildfire. Windows 95/98 simply lacks a real security model. Windows NT has a security model, but most NT configurations that I've seen do not take advantage of it (configuring it with real security in place takes away a good deal of normal Windows functionality). On the other hand, with a properly maintained Linux system, if one account is compromised, only that single account is compromised. The danger is contained; a mistake made by single secretary on a network of 15,000 cannot bring down the entire network and compromise all of the information; the loss is limited to at most what that one user has access to. Root kits are not terribly relavent to this discussion. They should not effect a properly maintained system any more so than DOS attacks or remote root exploits do. Under Linux, bugs that allow root exploits tend to be fixed within a few hours of their discovery. Piotr | ||
|