Blog::new

To follow up my previous post on svn/git : Linus has posted a pretty thorough explanation on why he thinks that decentralized SCMs work better than centralized ones. Contrary to his talk at google, this post actually convinced me of his point. Even though it doesn’t apply to all projects (not Rosegarden, for instance, since there aren’t enough developpers to justify this), I now understand what benefits a decentralized system brings in terms of flexibility, with almost no drawbacks indeed.

It really boils down to this : creating a branch and merging it anywhere (i.e. not just to the trunk) should be a trivial operation. Once you have this, branches become a much more useful and versatile tool, and you can start thinking your development from a different perspective.

Well, after months of wait and speculation, Canon finally announced the 40D. Specs look good, though I wish they’d look into a HDR sensor like the Fuji S5 Pro has. I don’t care about the full frame, my walk-around lens is the 17-55 f2.8 IS, for which there’s no equivalent in the EF line (not with IS). And given that Canon seems pretty committed to the EF-S lens line, it’s a safe bet the x0D line won’t be full frame anytime soon.

Aside of that, the real attractive feature is the WiFi grip. Along with the obvious file-transfer features, it also has gps connectivity and enables wireless shooting through http. Now that is geek-appealing :-).

Anyway, bottom line is, for me the upgrade is tempting, but not compelling. We’ll see.

I’m actually typing this while waiting for svn to merge a branch back to the trunk of rosegarden source tree…

Anyway, a while ago, I watched this talk by Linus at Google about his own Version Control System, git. I’ll skip the comments about his attitude and mention a few points I felt like commenting about.

1. When Linus says that KDE should have split its repository into multiple ones, I don’t see how that would make things any easier for devs. KDE has a base dependency, namely the KDE libs. How would you split KDE’s code into several repository while avoiding the problem of keeping them in sync ? Doesn’t that break the whole idea of a version control system ? “Hey, I need to backtrack kmail’s code to a week ago to check for a regression, now what version of kdelibs do I need to get it to compile, mhhh, let’s see…”. Yeah, that’s bright.
2. I wish Linus would show a basic workflow of how git works
3. And to the point of this post : Linus really does have a point when he says that a version control system really must make merging an easy operation. Yeah, even if branching is cheap (and remembering having to lock a cvs repository for hours just to make a tag or a branch, I can appreciate how important that is), if merging is a headache, it does defeat the whole purpose of it. That is, branches will still be seen as something you use only when you really have no other choice. And merging in svn really is a headache, simply having to find the version number of the branch’s origin is dull enough, but merge tracking is way too error-prone. Fortunately it should be part of the next version of svn, but I’m not sure leaving it out was the right decision. Then again they had to cut the feature list somewhere…

A couple of days ago, I made an order through Amazon which included a nabaztag. Despite what I still view as a design flaw, it’s still a cute, versatile toy and I figured I could find a nice use for it. Then, thinking more about it, I changed my mind and cancelled the order. The only real use I could find for it was to have it report some events like new mail or someone trying to reach me through IM, allowing me to still know about these events even when I’m not in front of my machine and the screen saver is on. But two things stopped me. First, the nabaztag itself not having a screen, displaying any kind of info means using some kind of colored light / ear positioning combination. It can actually be practical (red nose light => you’ve got mail), but doesn’t scale very well with the number of informations to report (”er, what was the two blue, one orange, left ear up sequence for already ?”). Second, well, do I really need all this info ? I think I already compulsively check my email enough as it is :-).

Let’s still indulge the fact that an external display unit would save me some time, knowing immediately if it’s worth checking for mail/news/IM contacts. Searching for an external LCD screen, the only one I could came up with was this one. Nice, but USB only, and not very legible from a distance.

Then, chatting about this with other geek friends, one mentionned the chumby. Ah, now this one really seems to fit my needs perfectly, except it’s not available yet. The widgets are actually written in Flash Lite, which should be good enough, so I think I’ll keep an eye on this one.

Finally, yet another option is… the iphone of course :-). Wifi, really mobile (as in “no damn power cord”), it’s actually probably the best option, even though it will probably require a bit more tinkering to make it do what I want due to its lack of openness.

In any case, I guess I have to wait :-).

I recently came across this article on Open Office and how buggy it is, despite the fact that it was Open Source (and therefore open to the scrutiny of thousands of hackers willing to help fixing it).

It perfectly illustrates the how ESR’s “enough eyeballs” are just about as mythical as the man-month.

The fact is, opening the source of any software is a requisite to attract hackers to help with it, but it’s by no means sufficient. After all these years dealing with OSS, I’m starting to think it can’t work at all except for a very small category of software. To gain a contributing community, software has to :

• be a hacker tool (like a kernel, a mail agent, a compiler - forget about business-related stuff, hackers don’t use spreadsheets or word processors)
• be easy to build (otherwise a would-be contributor will be frustrated before being able to do anything)
• be easy to find your way around (which means be modular, and business-related apps generally aren’t, because while it’s relatively easy to break down a kernel into specific modules, it’s much harder for this kind of apps)

Other than that, while opening the code still brings lots of benefits, you won’t get a community of hackers like the one of the Linux kernel.

Thankfully, the OSS community has matured a bit, and the idea that OSS isn’t the Silver Bullet doesn’t seem such a blasphemy anymore (except to newbies or non-coding zealots).

So, like many (’many’ being the understatement of the year here), I’ve watched the 20mn iPhone tour Apple has just released, and would therefore like to post my thoughts.

• if there’s a search feature for the contacts, I wonder why they didn’t show it. Otherwise this would be one area where my Treo is more convenient that the iPhone (typing a few letters instantly shows the contacts whose names start with the typed sequence). Given how large a contact list usually grows, searching through it alphabetically just doesn’t cut it, no matter how fancy the scrolling is. I really hope they haven’t pulled another ‘cursor keys are obsolete’ here (Jobs decided against adding cursor keys on the keyboard of the very first Macintosh, arguing that they were no longer needed now they had this wonderful new device called the mouse - I hope the iPhone’s ‘flick scroll’ hasn’t infatuated them the way the mouse did).
• It offers quite a bit of custom services (maps, voice mail) that will probably require some integration work on the side of the service providers. I hope they won’t have to trim it down because of this.
• the maps app really begs for a GPS device
• Does YouTube really need to be put forward as a major app ? As far as I can see, it’s still only a fancy time-waster, not a “serious” application. That it would be hyped like that really puzzles me.

On a side note : add me on the list of people who think that Ajax as a way to add 3rd party apps on the iPhone is a major disappointment. I just hope that, given how generic a name Ajax is, they didn’t just use any random javascript-based Ajax toolkit and went for a more XUL-like way of doing things. We’ll see.

Throughout my professional career (a bit more than 10 years now), I’ve worked with about 20 managers at various levels (from the team leader to the small company CEO). Among them, most have been between average and good, 5 were truly abysmal (i.e. actually counter-productive and detrimental to the project), and 2 were really outstanding. When I was a junior I used to think that managers were all representatives of the Dilbert Principle (and indeed I had seen at least one good example). As time made me a tad wiser, I found that not only management requires as much talent as development, but truly good managers are even more rare than truly good programmers.

So just for the sake of feeding this blog, here’s a short list of what I believe are key signs to a good manager (and to a bad one) :

• A good manager will remove obstacles and try to help you working more efficiently. He will keep his door open, ask what problems you encounter and will actually try to solve them. If the problem is yourself, so be it, but before reaching this conclusion, other solutions will be tried. A bad manager will rather coerce you into working as he thinks you should, and will not want to hear about your problems (the typical answer will be something along the lines of “if you can’t do it I’ll go find someone who can). By default, he sees you as a slacker (actual example : at the beginning of a worldwide sports event spanning several weeks, a company-wide mail sent by the CEO, reminding us that everybody should be there a 9AM and should not leave before 6PM, so don’t you people think of leaving to watch the game on TV, Big Brother is watching you).
• A good manager will follow what he says with actual actions. A bad manager may collect input from his staff, sometimes very thoroughly, but won’t act on it. (actual example : a 3h long meeting, with all recriminations scribbled on papers tacked on large boards. In the end, the manager just took a photograph of the boards, saying “I’ll study this”, and nothing ever happened).
• A good manager actually has a clue about which management techniques work and which don’t (generally comes through experience). A bad manager will blindly try to apply whatever buzzword he may have heard of, pretty much fitting the “if your only tool is a hammer, all problems look like nails” stereotype.
• A good manager will actually welcome comments and remarks on his decisions, even negative ones. He will accept debate and will even expect it, even though he will have the last word, because that’s his role. He won’t have a problem with recognizing he is wrong, as he knows he can’t always be right. A bad manager will feel threatened by differing points of view, and will perceive them as a challenge of his authority. Basically, a good manager is self-secure, a bad manager isn’t, and tries to prove otherwise by abusing of his authority.
• A good manager tries to entice good behavior, a bad manager only punishes bad one. For instance a good manager will publicly praise you for an achievement. A bad manager will publicly scold you for a mistake (actual example : a manager proudly exhibiting at every weekly team meeting a chart showing the developers with the highest number of assigned bugs).

Skimming through a blog about the Cannes Film Festival, I came across this movie website URL. Movies have long had web sites, but a myspace page ? This is new.

Let’s be honest : myspace is the new geocities. Back in the early days of the web, geocities was a huge heap of ugly, boring personal web pages, generally very badly designed and using every single dumb javascript trick in the book (that, and the <flash> tag). It was the place where bright pink text on yellow background would wrap around blurry holiday pictures scattered at random. And your mouse cursor would be followed by a trail of cute little stars. Geocities was the haven of thousands of vanity sites. Now doesn’t that make you think of something ?

Why was geocities so successful ? Because it offered free and easy to use hosting. Forget about quality, the “worse is better” principle applies. Why is myspace so successful these days, even though just about 99% of its webpages are eye-piercing demonstrations of ugliness ? I can’t see any other reason that it’s because it gives all the tools a home site needs in a nicely wrapped package : blog, photo gallery, audio and video playback, and… hyperlinks. Because, really, what are “friends” in myspace if not a reinvention of hyperlinks within the realm of myspace ? Yes, hyperlinks within a text are used to offer a way to “more info”, but in a blogroll for instance they are an endorsement. Much like myspace “friends”.

Then of course there’s also the critical mass factor, which myspace has reached long ago. It’s sheer weight makes it mandatory for media marketing droids to set up camp here whenever they have something new to promote. And so myspace becomes a fac simile of the Net, except centralized.

This is actually a global trend as Nicholas Carr says. First were centralized, disconnected networks (AOL, Compuserve, Genie… my first online steps where on a system called Calvados, which later became Calvacom). Then Internet wiped them out. And now we see them growing back, within the Net.

It seems that I wasn’t the only one making the connection with the open source ecosystem.

How long do you think an email address can stay in the spam lists ? I just quizzed a bunch of friends about this, one immediately answered “indefinitely”. He’s right.

Back in 1992-93, my first job was being a trainee at the INRIA. There was no web back then, we actually witnessed its birth (and all we did was shrug). Usenet was the main channel for discussions, and I used to post in a few newsgroups. Spam was exceptionnal, people doing it (generally college kids who didn’t know better) were rebuked by people receiving their messages and would even apologize when they would realize what they had done.

By some strange twist of fate, I find myself working again at the INRIA today, some 15 years later. I started on april 2nd. When I first logged on my office machine and looked at my mail, I was very surprised to find about 45 spams in it. Most of them were directed at my own address, rather than at one of the mailing lists I had been subscribed to by default.

I was rather surprised but given the amount of spam we get these days, I figured it was to be expected, since it couldn’t be possible that this was because my address was a “once valid” one (since the INRIA has kept the same policy for creating logins and mail addresses, in effect they had “ressurected” an email address that used to exist 15 years ago). Or so I thought. Then I noticed that some of this spam was directed at a host of the inria.fr domain that couldn’t possibly be “seen” from outside. It was the host I used to post to Usenet from.

So it really goes like this : create an email adress, post on Usenet or somewhere where it will be harvested, disable the adress, wait 15 years, enable the adress… and watch the spam pour in.

Yuck.