Content-Length: 36144 | pFad | http://lwn.net/Articles/718043/#Comments

Eudyptula Challenge Status report [LWN.net]
|
|
Subscribe / Log in / New account

Eudyptula Challenge Status report

The Eudyptula Challenge is a series of programming exercises for the Linux kernel. It starts from a very basic "Hello world" kernel module, moves up in complexity to getting patches accepted into the main kernel. The challenge will be closed to new participants in a few months, when 20,000 people have signed up. LWN covered the Eudyptula Challenge in May 2014, when it was fairly new. At this time over 19,000 people have signed up and only 149 have finished.


From:  "Little Penguin" <little-AT-eudyptula-challenge.org>
To:  <twoerner-AT-gmail.com>
Subject:  Eudyptula Challenge Status report, March 2017
Date:  Mar 24, 2017 4:33 AM

Hi,

Welcome to another very semi-irregular update from the Eudyptula
Challenge.

Just a short update this time, in bullet format:

        - challenge has been running for over 3 years
        - over 19000 people have signed up for the challenge
        - only 149 have finished it
        - that percentage is really low :(
        - little is getting tired
        - challenge will be closed to new people once we reach 20000, go
          tell your friends to sign up if they wish to participate.
        - all queues are empty, response times are fast, what is keeping
          you all from finishing?

Don't worry, the challenge will live on after it stops accepting new
people (at the current rate, in a few months), it just will look a bit
different...

- little


to post comments

Eudyptula Challenge Status report

Posted Mar 25, 2017 13:12 UTC (Sat) by geuder (subscriber, #62854) [Link] (12 responses)

Certainly interesting. But why should I deal with someone hiding behind a pseudonym? The kernel doesn't accept patches by pseudonyms either.

Eudyptula Challenge Status report

Posted Mar 25, 2017 14:24 UTC (Sat) by adirat (subscriber, #86623) [Link]

Do you ever have fun while programming?

Eudyptula Challenge Status report

Posted Mar 27, 2017 8:57 UTC (Mon) by lkundrak (subscriber, #43452) [Link] (6 responses)

I am not sure if I understand you correctly. Do you ask how do you benefit from the challenge?

If it's not obvious; you (hopefully) get to learn something new. You'll have fun. You end up being more knowledgeable. It might widen choices of what you end up doing for living.

How is the Little Penguin's pseudonym relevant in that context? I find it rather cute.

Eudyptula Challenge Status report

Posted Mar 27, 2017 14:32 UTC (Mon) by geuder (subscriber, #62854) [Link] (5 responses)

> Do you ask how do you benefit from the challenge?

No.

When doing the challenge I obviously interact with someone. I would just feel more comfortable if the someone mentioned his (unlikely to be her...) real name.

> How is the Little Penguin's pseudonym relevant in that context? I find it rather cute.

I don't care what kind of nickname they like to use, cute or not. Fine with me as long as the real identity weren't hidden.

Eudyptula Challenge Status report

Posted Mar 27, 2017 14:56 UTC (Mon) by Wol (subscriber, #4433) [Link] (4 responses)

How do I know "geuder" is not a pseudonym? Doesn't sound like a real name to me :-)

Oh - and "Wol" sounds like a pseudonym, but it's actually an (unusual) shortened form of my real name :-) (Which is why I use it.)

Cheers,
Wol

Eudyptula Challenge Status report

Posted Mar 27, 2017 18:41 UTC (Mon) by geuder (subscriber, #62854) [Link] (3 responses)

> How do I know "geuder" is not a pseudonym?

You don't. If you were a subscriber you could filter me, if you didn't want to read comments from some usernames, e. g. those you find suspicious. (At least I understood so, have never used any filtering on lwn.)

> Doesn't sound like a real name to me :-)

That's your problem:) That's the way it is spelled in my passport, at least if comparing case-insensitively.

(Actually had I known that the name is displayed in discussion forums when registering, I might have chosen something else: In forums that require a full name I use it. Otherwise I use a nick.)

Anyway, what I was referring to:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/...

The rule might be there for legal reasons. But in the end no trusted chain of PGP signatures is required, so mostly it's a question of personal integrity whether people really use their real names. I'd say at least 95% of the open source world works like this.

It's just a matter of style and trust. In the common sense, not the cryptographic one of course.

In the end, what I would have expected from my post is (maybe) a good argument, why a presumably well known kernel hacker would do this effort without telling his name. None has come up so far as far as I can see, so we should stop posting less and less informative follow-ups. At least I'll try to do so unless a novel viewpoint comes up.

Eudyptula Challenge Status report

Posted Mar 27, 2017 20:29 UTC (Mon) by ballombe (subscriber, #9523) [Link]

Given the number of applicants, it is more likely to be an alias for a large group of people.
Would you be more comfortable if the email was signed 'The Eudyptula challenge team' ?

Beside, it is customary to withhold the names of the referees of scientific papers. Maybe similar rationale apply.

Eudyptula Challenge Status report

Posted Mar 28, 2017 16:20 UTC (Tue) by adirat (subscriber, #86623) [Link]

Here's a novel idea: the issues you have are solved by medicine, not argumentation.

Eudyptula Challenge Status report

Posted Mar 29, 2017 14:45 UTC (Wed) by snits (guest, #37976) [Link]

Maybe they didn't want their email and other means of communication flooded with questions and requests?

Eudyptula Challenge Status report

Posted Mar 27, 2017 16:09 UTC (Mon) by Kamiccolo (subscriber, #95159) [Link]

Are You sure it's a pseudonym? ;)

Eudyptula Challenge Status report

Posted Mar 30, 2017 11:56 UTC (Thu) by azz (subscriber, #371) [Link] (2 responses)

> The kernel doesn't accept patches by pseudonyms either.

Yes, it does (in practice):

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/...

Spelvin

Posted Mar 30, 2017 12:49 UTC (Thu) by corbet (editor, #1) [Link] (1 responses)

So I asked around a bit about this particular case a while back. The real identity of this particular developer is known to the people who take their patches. So it's a pseudonym, but a special-case pseudonym that is not 100% opaque.

Spelvin

Posted Mar 30, 2017 14:32 UTC (Thu) by farnz (subscriber, #17727) [Link]

Out of curiosity, is anyone able to disclose why said developer uses the George Spelvin pseudonym? If the reasons are too personal (or too revealing of Mr Spelvin's hidden identity), I understand, but my curiosity is getting the better of me here.

Eudyptula Challenge Status report

Posted Mar 25, 2017 18:12 UTC (Sat) by fratti (guest, #105722) [Link] (8 responses)

For what it's worth, I think 149 people who have completed the challenge is still a sizeable number of new kernel programmers, assuming those weren't just existing kernel contributors doing the challenge.

Eudyptula Challenge Status report

Posted Mar 26, 2017 7:42 UTC (Sun) by lkurusa (guest, #97704) [Link] (7 responses)

There's a fair amount of people I know who started doing the Eudyptula challenge and had stopped in the middle to start working on the kernel full-time, so that number maybe a lot higher.

Eudyptula Challenge Status report

Posted Mar 26, 2017 20:34 UTC (Sun) by jmichels (subscriber, #98352) [Link] (5 responses)

I started working on it for fun. I'm still working on it, but a few issues have made it less of a priority for me.

First, I've had submissions sit in the queue for a very long time. 6 months was the worst.

Second, I've had too many submissions rejected for not doing it the way one way that little expects. I don't mind doing it the way little wants. The problem is that little doesn't tell you this ahead of time, so it kind of turns into a guessing game. I also get that the point of this is to expose you to different parts of the kernel Api in a guided way. But the guessing, coding, testing, and submitting becomes time consuming and very unsatisfying. Especially when you submit a correct and functional answer only to have it rejected for not being the correct functional submission that little wants.

Eudyptula Challenge Status report

Posted Mar 27, 2017 9:09 UTC (Mon) by lkundrak (subscriber, #43452) [Link]

Thanks for sharing your experience. I think mine was a bit different. I finished the challenge a long time ago, so I probably don't remember some details and some circumstances were likely different -- perhaps queues were shorter and Little was a bit quicker to answer.

To me it seems that the slowness of responses and rejections of solutions that work but still need some improvements teaches you patience that is so essential in collaborative software projects. The responses were always rather good; even though sometimes a bit challenging to understand.

As for me I've had great fun doing the challenge and recommend it to everyone. But arm yourself with patience and dedicate time to it, especially if you're not a veteran in the area.

Eudyptula Challenge Status report

Posted Mar 27, 2017 14:50 UTC (Mon) by geuder (subscriber, #62854) [Link] (3 responses)

> Especially when you submit a correct and functional answer only to have it rejected for not being the correct functional submission that little wants.

They say it's a bunch of scripts behind the scenes. So maybe too little human review and as soon as you code a bit different from what the scripts' author expected you will get a reject? Of course formatting that checkpatch.pl does not like should be rejected. But even when checkpatch.pl is happy every programming problem has an infinite number of valid solutions (or at least a very big one for all reasonably practical cases)

Just guessing, no experience.

Of course it happens all the time in development teams, whether open source or not, that the reviewer has some points that the author does not agree with. Either deeming the reviewer's suggestions completely equivalent to the code under review or even worse.

Eudyptula Challenge Status report

Posted Mar 27, 2017 17:52 UTC (Mon) by ntnn (guest, #109693) [Link] (2 responses)

I'm doing the challenge at the moment and I have the feeling that it actually is a human/some humans doing the review - just with the help of a lot of scripts, so their input would be nothing more than 'reject' or 'accept'

E.g. I send in a solution yesterday night and was at place four. I got the next task today around midday - automated scripts would work a lot faster.

Eudyptula Challenge Status report

Posted Mar 28, 2017 2:41 UTC (Tue) by rav (subscriber, #89256) [Link] (1 responses)

I have the same impression. I advanced from Task 2 to Task 5 over the weekend (after having completed Task 1 almost three years ago). Grading Tasks 2 to 4 took 15 hours, 4 minutes and 19 hours, respectively; I received the grade from Little at times Sun, 26 Mar 2017 08:47:23 +0000, Sun, 26 Mar 2017 19:26:21 +0000 and Mon, 27 Mar 2017 15:44:12 +0000.

I'll submit Task 5 tomorrow (when I have access to a USB keyboard, he he he... I've probably revealed too much now).

Eudyptula Challenge Status report

Posted Mar 28, 2017 10:05 UTC (Tue) by JacobvonChorus (guest, #109772) [Link]

I just signed up, and look forward to getting my hands dirty. It's a neat initiative, I'm a strong believer in learn-by-doing and this seems really good.

Eudyptula Challenge Status report

Posted Mar 30, 2017 2:37 UTC (Thu) by ajdlinux (subscriber, #82125) [Link]

I'm one of them - didn't get through Eudyptula before I ended up being paid to work on powerpc.


Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds









ApplySandwichStrip

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier!      Saves Data!


--- a PPN by Garber Painting Akron. With Image Size Reduction included!

Fetched URL: http://lwn.net/Articles/718043/#Comments

Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy