Open Source

Intercepting Bug Reports is Bad

open sourceParticipating in and maintaining open source projects is great fun, much rewarding and very educating. One thing you always want is bug reports from users who suffer from problems, as you cannot fix problems unless you know they exist!

Yet there are several obstacles along the way that can prevent users’ reports from reaching your project. These obstacles include:

Security sites

Eager to announce a new problem, a new revealed leak or exploit, people (often) submit security- related problems directly to sites and forums dealing with security. These sites (of course) don’t forward the reports onwards, they simply assume the projects are informed as well…

Distros

People who use Linux Distributions very often feel like a user of that distro (no surprise there really!) and they therefore primarily report bugs and problems to the distro’s bug tracker. Unless the people in the distro are keen and interested enough, those reports sit there rotting away and people in the upstream project who would like to know about it are never told and thus the bug isn’t fixed…

Sometimes the bug is even fixed by the distro people and they make a newly built version available, featuring that patch, but the patch isn’t forwarded upstream either!

Related forums/mailing lists

People discussing the project in another list or forum where they are users of it. They talk about workarounds etc and sometimes even talk about “known bugs” and “existing flaws” but without ever reporting them to the originating project so they aren’t fixed. They may thus be known to the subgroup there but not upstream.

Please report upstream!

This is my cry for how this situation can be fixed: make sure that problems you know of are reported upstream to the actual project working on the project. Don’t assume that reporting it to your distro or to your neighbor is enough!

(I could easily point out examples for all these cases for projects I am involved in, but I don’t think pointing fingers will gain us anything.)

cURL and libcurl

curl and libcurl 7.17.1

7.17.1 – the 102nd release of curl is out, with less than 5 months left to our ten year anniversary!

The previous release (7.17.0) included a few larger internal changes and unfortunately that had the backside that it brought a whole array of new bugs in, that we now have spent almost two months polishing off.

cURL

Apart from the twenty or so bug fixes, a range of new things are introduced as well, including improved NSS support, –proxy-negotiate, –post301 (to make curl act more standards compliant on HTTP 301 responses), –hostpubmd.

libcurl hackers will appreciate CURLOPT_OPENSOCKETFUNCTION and CURLOPT_COPYPOSTFIELDS (the latter a complement to the existing CURLOPT_POSTFIELDS that got broken in 7.17.0 if you posted binary data that contains a zero byte).

7.17.1 contains contributions by at least 16 different people (me not included).

Rockbox

Rockbox on iPod Touch

iPod TouchWell, not really but at least the recent “jailbreak” for it opens up the possibility…

The jailbreak seems to open up the ability to run apps on the target that are built for it, so I figure it can then theoretically be used to run whatever, and that’s why I say it is an opening for an eager and dedicated person to get Rockbox going on it.

cURL and libcurl

curl references

I amuse myself by occasionally reading up on articles and posts “out there” that talk about curl and libcurl, and I often find interesting snippets and data well worth reading. Here’s a few of the ones I’ve stumbled upon recently:

  1. Ryan Boyd of Google posted an article on how to use curl to interact with Google’s data apis.
  2. Tony G wrote a post to a u2 database mailing list and did an indirect praise of curl.
  3. magnetk.com writes about how to build a recent libcurl with visual studio 2005
Linux

Distros Don’t Drive Development

Lots of press and people focus on Linux distributions when they check out what happens in Linux land. This and that distro come in new releases and they offer this and that brand new feature. This is also true of the many linux podcasts. They give credit to distros for new things that pop up.

Mostly everyone of us involved in open source projects have a different view on all that:

Distros package the developments that are done elsewhere by other people. Sure they contribute with glue code and they do put pieces together in useful ways but really, most of the real grunt work – the actual sweating, is done by ordinary open source teams working independently of distros (but sure, sometimes people related to or even employed by distros contribute in such projects).

This is also part of the explanation to why most distros are at about the same level of development, why no distro outcompetes the others at one go. When X11 brings a fancy new feature, all distros have it. When compiz can rotate the screen in yet another way, all distros ship that…

Of course the other part of the explanation is that most distros release their own work as open source, free for the other distros to absorb.

In fact, many times the distros actually hinder the work of the open source projects since they add a filter to bug reports, they patch in their own dirty solutions in their distro rather than to work with the projects to do the fix the best possible way and similar.

Distros are exactly what they’re called: distributions – they distribute bundled collections of software, the vast majority of that software is not made by the distro.

IT Politics

I Like GMT+2 better

1 Comment

In Sweden we switch to good time (daylight wise) in the early spring and we switch off the good time over to bad time in late October (the 28th this year).

Seriously, the non-daylight savings time is not good for us not even during winter. We’d be much better off with GMT+2 all the time.

I’m not sure this is because we’re this far up north in this particular time zone (Central European Time), or if perhaps most citizens in it feel the same. And if they do, would that imply that most of the people in neighboring zones also think they have the wrong time?

cURL and libcurl

wget going libcurl?

Micah Cowan is the current maintainer of GNU Wget, and he recently posted a long mail to the wgetA gnu head! mailing list titled “Thoughts on Wget 1.x, 2.0“.

Two fun quotes for the curious who don’t feel like reading the whole post:

1. On the subject of making wget deal with multiple simultanous connections/requests: The obvious solution to that is to use c-ares, which does exactly that: handle DNS queries asynchronously. Actually, I didn’t know this until just now, but c-ares was split off from ares to meet the needs of the curl developers.

2. Following the first reasoning, they can indeed get away with even less work if they base that work on an existing solution: While I’ve talked about not reinventing the wheel, using existing packages to save us the trouble of having to maintain portable async code, higher-level buffered-IO and network comm code, etc, I’ve been neglecting one more package choice. There is, after all, already a Free Software package that goes beyond handling asynchronous network operations, to specifically handle asynchronous _web_ operations; I’m speaking, of course, of libcurl. There would seem to be some obvious motivation for simply using libcurl to handle all asynchronous web traffic, and wrapping it with the logic we need to handle retries, recursion, timestamping, traversing, selecting which files to download, etc. Besides async web code, of course, we’d also automatically get support for a number of various protocols (SFTP, for example) that have been requested in Wget.libcurl

I am of course happy to see that the consideration exists – even if this won’t go further than just expressed in a mail. I did ventilate this idea to the wget people back in 2001, and even though we’re now more than six years down the road since then the situation is now even more clear: libcurl is a much more capable and proven transport layer solution and it supports much more protocols than wget is/does.

Me biased? naaah… 🙂

Electronics

Tomtom MapShare

1 Comment

Tomtom ONEI have one of these lovely little Tomtom ONE GPSes, and the other day my brother notified me about their cool recent update to their firmware!

Now they feature map correction abilities, and apparently your map corrections and added details such as points of interests etc can be shared with your friends!

Very cool (sounding). This turned out to be a little easier said than done (as usual). First, I had to boot my computer into in Windows and run the “Tomtom HOME” software. It turned out to crash repeatedly after about 20-30 seconds of use, but after a lot of restarts it managed to upgrade both the GPS unit and get a fresh new “HOME” software version. With the new HOME software it stopped crashing and funnily enough the new version immediately downloaded yet another new HOME upgrade…

Anyway, now I have the latest firmware and I try out the menu option for map correction only to see… that it claims my maps are too old to use this feature! OK, I have the “Scandinavian” version and using the Tomtom HOME it says I already have the latest version… I assume this means that I should rather buy another map set or similar. It claiming I have the latest is suspicious anyway since I know my brother have newer maps. After some checking, I learned that he bought newer maps a while ago. Ok, time to get updated maps. In fact, they don’t even sell “Scandinavian” maps these days so I guess my map route was a dead end, now the closest thing is “Sweden, Norway, Denmark, Finland” which is almost twice as much data – and thus such an upgrade requires me to get a larger SD card first! 😉

I’ll get back when I’ve got a new card and a new map!

As many times before, one can’t but to wonder what my parents or similarly (not at all) tech-skilled persons would do if they would face this kind of challenge…

cURL and libcurl

curl work towards release #102

libcurlThis is a bug fix week in curl land, trying to get everything sorted and fine to be able to release a really fine 7.17.1 release within a week or so. We got some nasty memory-related problems with changed protocols re-using the same easy handle, but it was good that they crept up and I think we’re doing good changes now that stabilize curl.

Release date now targeted perhaps around October 28-30.

curl, open source and networking