[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Projects in the Hurd

From: Tom Hart
Subject: Re: Projects in the Hurd
Date: Wed, 29 Jan 2003 10:00:18 -0600
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20021130

I've been looking at the Hurd for my undergrad topics course.

One question my Prof has posed to me is, "How could a program like Apache, which is highly-threaded and servicing multiple HTTP requests, bet better implemented under a multi-server system such as GNU/Hurd?"

In addition, we've found looking forward to L4-Hurd to be quite interesting, since Dr. Liedtke had some very good ideas about microkernels. Also, the Hurd community, along with other multi-server projects (ie. SawMill and icky-proprietary QNX) have a lot of documentation on the advantages of multi-server systems.

We've also been thinking about how the code for utilities like APT could be trimmed in the presence of a system that supports translators; then, APT wouldn't need any code specific to FTP/HTTP/FILE/CD-ROM, since the translators could be the only part of the system that see the difference.

W.r.t. programs with translators, of course, this (superior) style of programming couldn't come about at present, since people would still want APT to work on systems based on Linux and *BSD kernels. Something I've been thinking is, I don't see how translators are inseperably tied to multi-servers. Couldn't a Linux or *BSD kernel be hacked to support translators?

Now, maybe I'm barking up the wrong tree here, but it seems to me that if translators could be ported to these systems, then: 1. It would make it easier for programmers to move to a Hurdish style of programming while the Hurd is still stabilizing; 2. The Hurd would get good publicity, since the translator mechanism is a very good idea, and it came from the Hurd developers; and 3. There would be a bigger pool of people helping to solve the problems translators introduce (I'm thinking of discussions on the mailing lists involving UnionFS, for instance).

Well, that's my brain dump. I hope it gives you guys some stuff to think about. Good luck with the dissertations! =)

Bharata B Rao wrote:

We are looking for projects for our MS dissertation course. We want to know if 
there we can do something in the Hurd which will be useful. We went through the 
tasklist at savannah, but we are not sure if it is uptodate.

Could you inform us about any significant things that need to be done in the 
Hurd, which we can take up ? We have written code for Linux kernel and have 
reasonable understanding of the Linux kernel.

Please CC the replies.

   /                                               |
  /  Tom Hart                                      |
 |   hartte13@BrandonU.ca                          |
  \  "rmTFM - Build consistent interfaces."        |

reply via email to

[Prev in Thread] Current Thread [Next in Thread]