Archive for November, 2005

Google Analytics

Is there anyone out there who has been able to get any reports out of Google Analytics? I signed up yesterday morning and after about 28 hours it’s still saying: “Analytics has been successfully installed and data is being gathered now. Your first reports will be ready within twelve hours.”

I was looking forward to getting rid of StatCounter, MyBlogLog and PHP-Stats. They served me well until now, but if I can get everything from a single source and it’s free too, why not?

Having to wait is particularly frustrating now that, due to a certain video, my visitors have just tripled overnight.

Dilbert on Evolution

Scott Adams: “To me, the most fascinating aspect of the debate over Darwinism versus Intelligent Design is that neither side understands the other side’s argument. Better yet, no one seems to understand their own side’s argument. But that doesn’t stop anyone from having a passionate opinion.”

Scott’s blog entry exemplifies some of the most common misconceptions people have regarding the Intelligent Design vs. Evolution debate.

icon_dilbert.gifFirst misconception: ID and Evolution are somewhat equivalent as “theories” since both have plenty of “holes” in them, so we cannot really decide which one is better at explaining reality and we’d be better teaching them both. There are two errors here. The first one is that ID really has no holes: Whenever someone points out a hole in it, the obvious response is “Someone did it!”, where “someone” is an unnamed, all-powerful designer which the ID-ers, were they not the bunch of hypocritical liars they are, would call “God”.

Second: Of course Evolution theory does not explain everything regarding evolution with 100% certainty. This does not make it less of a scientific theory. All scientific knowledge is tentative, partial and provisional. When a scientist has not enough data to provide a probable explanation, he might put forth a testable hypothesis. An ID-er will just say: “Since my limited imagination cannot understand how this came to being, I will simply postulate that someone did it.” This is not science.

th_FSM3d.jpgSecond misconception: “The Creationists and the Intelligent Design folks have the same target (Darwin), but they don’t have the same argument. The average person who has a strong opinion on this topic doesn’t understand that distinction because the political agenda of the creationists makes things murky.”

Intelligent Design is just creationism in disguise. Maybe not of the young-earth, noah-flood-did-really-happen kind but creationism it is. The political agenda is the same: Discredit Evolution and put back religion in the science classrooms and books. It is just so evident.

In the end, the deciding factor between ID and Evolution is not whether one is “truer” than the other one. For all we know, the truth might be that the Flying Spaghetti Monster created the world last Thursday, with the appearance of an old, evolved world. You simply cannot test this hypothesis, just like you cannot test ID’s designer. Therefore, it is not science and shouldn’t be taught in a science classroom. Is this so hard to understand?

Update: PZ Myers (a scientist) chides Adams much better than I could ever hope to.

Rails not ready for the enterprise?

So I’m starting to learn something about Ruby on Rails. I’m not actually doing anything at the moment, as I have a tight deadline looming and very little time for playing, but I ordered the “Programming Ruby” and the “Agile Web Development with Rails” books from The Pragmatic Programmer bookshelf.

Meanwhile, I’m keeping my eyes open, reading blogs and lurking on the mailing list. From blogs I caught these two snippets in the last days:

Berin: “Ruby and Rails both have a wonderful programming environment. It is fun to create using the language. However, where it does fail is when you have to deploy on the server side. I could run it as a CGI script, but it is slow. I could use FCGI, but integration with Apache2 is shaky at best and I get unexplained errors between development and production environments. I can’t seem to use mod_ruby, and no amount of searching, asking, or whining seems to get me the help I need. This is the greatest hurdle to Ruby on the server. It is reasonable to believe that the Ruby community can make this happen. I mean what happens when you can’t use LigHTTPd and FastCGI? There just needs to be more people helping out.”

Andrew: “Then I lost a day and a night to the joys of Ruby on Rails on Debian. Ready for prime time? Not without a LOT of pain. This is one framework that needs to grow up before I’d recommend it for general use; you need to be a genius sysadmin and no more than one step removed from divinity to have any hope of getting it to behave on a production server. More on that later, just as soon as I can write about it without letting out blood-curdling howls of anger and pain.”

I’ve also been following the mailing list thread regarding internationalization support in Rails, which seems to be not quite there yet. There are solutions, but apparently none very complete and stable.

So, Rails might not be as enterprise-ready as Java, but I’m not really worried about this. It’s a young framework that hasn’t released its 1.0 version yet. Do you remember what Java was at version 1.0? I remember starting to learn and like Java even before 1.0 came out and this investment surely paid off in the end.

Moreover, as I wrote above, it’s going to be some time before I think about developing some serious apps with it. This will give Rails time to mature a little bit more.

Lapo ad

Don’t even try to understand this if you’re not Italian and know about recent events concerning Mr. Lapo Elkann, but this is just too fun. If you search on Google for “lapo” you’ll get the following ad:

Lapo ad.png

New incredible Ronaldinho video

Ronaldinho 2.pngFrom my webserver logs, I know that lots of people are coming here from search engines, doing queries like “ronaldinho video” just because some time ago I posted here a link to a video of Ronaldinho scoring a fantastic goal against Chelsea in last season’s UEFA Champions League.

Today I want to give those people some more fun stuff to download. Here’s a video (from Nike’s site) of Ronaldinho doing some incredible things with a football. So incredible that many believe the video to be a fake, computer generated or such. I have some doubts myslelf but if it’s really a fake, it’s very good.Well, judge by yourself.


Forty Faces: A suggestion

I predict that Forty Faces’s success — if it does succeed — will bring its own failure, ultimately. As the number of scanned blogs grows, each blogger’s face will be present on the page for a much shorter time, which will void the basic usefulness of the site: catering to the bloggers’ vanity (including yours truly).

In order to delay the inevitable, I want to suggest Philipp to implement a small change: don’t let each blogger’s face appear more than once in the homepage. Having multiple mugshots of the most prolific bloggers like Scoble or Cory is not fun, after a while. Other bloggers who post less often, but write longer (and maybe more thoughtful) posts, don’t deserve to be pushed off the page so quickly.

Amazon Gets Patents on Consumer Reviews

I’m a bit worried by this:

Susan Kuchinskas, Internet News: “In one embodiment of the patent, the system sends consumers a message inviting them to write a review in a predetermined amount of time after the purchase. It’s a method widely used by online retailers, including Yahoo Shopping. The patent also covers the method of tracking who returns to rate products by asking them to click on a unique link in an e-mail.”

Apart from being worried in a general way about the ongoing spread of patents concerning business methods, I am worried in particular because when I designed The Open Source Zone, one of the explicit design goals was to build an Amazon type storefront, with user contributed reviews at its center.

Now, we don’t send messages to users inviting them to write reviews, but Susan also writes: But the patent even covers collecting reviews by letting visitors to a Web site fill out a form., which seems to cover also our usage. I really hope this is not the case.

I’d also like to know whether the patent is enforceable outside the U.S.

(Via Matt Asay.)

Forty Faces

fortyfaces.gif

Philipp Lenssen: “My new site is live: Forty Faces.com. Go have a look, it ought to be self-explanatory!

So what does the site do then? Whenever a blogger updates his or her post, a portrait of that blogger will be shown top (there’s a delay as RSS shouldn’t be polled too often). Older posts, and their accompanying faces, move down the page over time. 40 faces will be shown at any time, but the number of bloggers participating is not limited.”

I’ve sent a request to be added, so you will probably see my face there, sooner or later.

Opera Mini

ScreenShot2.jpgI was looking for a decent browser for my Nokia 6630, when I learned that Opera Software has released Opera Mini for free and immediately grabbed it. At first sight, it looks good and certainly many times quicker than the stock Nokia browser. The problem is that I’m quite cold towards mobile web browsing, so I don’t think I’ll be using it a lot, but still it’s nice to know you have an alternative when you need it.

Another problem is that it shows the main content column of this site with a black background. Not exactly readable!

“The Best Software Writing” review

1590595009.01._AA_SCMZZZZZZZ_.jpgThe Best Software Writing I: Selected and Introduced by Joel Spolsky

The blurb on the back cover reads:

At my own company, we instituted a policy: we only want to hire software developers who can write, and write well. Just try submitting a résumé to me with grammatical errors or a cover letter that reads like it was written by that guy who translates those instruction manuals for really cheap consumer electronics. I dare you.

I think this is a great policy. Being able to communicate well is one of the greatest and most overlooked assets to look for in good software developers. With this book, Joel Spolsky shows us some of the best examples of good writing in the field of software.

If it were just a matter of good writing style, the importance of this book would be rather limited. But its real value, besides the stylistic aspect, is the great content. Most articles are not just well written, but also very interesting, insightful or outrightly funny. Great content in a nice package.

I’ve previously referred to Paul Graham’s essay “Great Hackers”, which is one of the best. My other favorites are Ken Arnold’s “Style is Substance” and why the lucky stiff’s introduction to the Ruby language.

My only small gripe with the book is the fact that most, if not all, the material inside is already available on the net. It sure is pleasant having it all nicely collected and printed in a book, so that you can read it on the plane or at the beach, but I find it quite improbable that all good software writing can be found on the net and none in print magazines and book. Or maybe it is simply easier to obtain republishing rights for online material than it is for printed stuff? A couple of comic strips seem quite out of place, too.

Apart from these minor details, a very entertaining and stimulating read. Highly recommended.

Google Local for Mobile

Matt Croydon: “There goes Google, taking it to the next level again. Their latest offering is Google Local for Mobile. What they’ve done is essentially taken the Ajax model for desktop browsers and brought it the phone using J2ME. The idea is to have a fairly lightweight MIDlet (38k for the MIDP2 version) that then grabs all the data neccesary from the network.”

ScreenShot.jpgActually I don’t know what’s this got to do with Ajax, since it is apparently a Java midlet and has nothing to do with either Javascript or XMLHttpRequest, but it’s a cool app anyway. Google never ceases to amaze!

A couple of notes.

I was able to install it on my Nokia 6630 by specifying “Nokia 6682″ as phone model. My provider is Vodafone IT, so I choose “Other” as provider.

Even though the level of detail it offers for Italy is nowhere near useful, I was astonished to learn that it is actually higher than what is provided by Google Maps, as you can see by confronting the screenshot above with the corresponding area on Google Maps.

I can’t wait for the day when this type of service will have enough data to be useful here too.

Weasely Design

Advocating the teaching of Intelligent Design in schools seems to be the weaseliest possible behavior, al least according to Dilbert readers.

Dilbert.com - Weasel Awards:

Weaseliest Behavior

Advocating the teaching of Intelligent Design in schools
9661 
Gas price gouging
7828 
Reporting it as “finding supplies” when white people loot
6157 
Corporate boards approving CEO pay packages
6149 
Politicians blaming other politicians
6131 
Outsourcing
2687 
Downloading music or movies without paying
864 

Memeorandum widget

Gabe Rivera: “Want to include top memeorandum headlines in your blog sidebar?

Just use the following Javascript snippet:

<script language="JavaScript"
  src="http://memeorandum.net/pwidget-gray.js"
  type="text/javascript"></script>"

Done, and it looks cool, even though I don’t know how useful it can be, tucked there below the fold (scroll down, baby) in an already too crowded sidebar. Probably it’s more useful in a different kind of site.

Also, I don’t like to call it an Internet Connected Component (ICC), as proposed by Scoble. “Widget” is fine with me. Most people will instantly recognize what a widget is, but yet another TLA? Gimme a break!

Hackers, Open Source and business models

Paul GrahamI was recently re-reading Paul Graham’s classic piece “Great Hackers”, the reason being that Joel Spolsky included it in his “Best Software Writing” book, which I’m currently reading.

I have to admit that, when first I read it, it left me a bit cold. I particularly resented the opinion that no great hackers would autonomously decide to program in Java, when presented with an alternative like Python. I thought, and I still think, that the incredibly rich variety of Java Open Source projects couldn’t have grown out of dilbertesque corporate types. There truly must be some great hackers behind the best Java software. I even think I might have met some of them, unless the real great hackers are from another planet entirely.

Language issues aside, Paul’s essay was widely criticized at the time for being totally out of contact with reality. By staffing a company with hackers, the critics said, you will never get anything done. By letting them work on just what they like, they’ll never work on the customers’ problems.

Possible, but apparently companies that are doing more of this, like Google, are more successful than those that do less of it.

Nowadays, after having re-read “Great Hackers” and some other of Paul’s essays (in particular “What Business Can Learn from Open Source”), I am feeling more and more resonating with his opinions.

One of the things I’m starting to be convinced of is the fact that the killer applications of the next generation won’t be written in Java. Surely, they won’t be written using any one of the numerous, complex and bloated Java frameworks that litter the landscape. They’ll be written using Ruby, Python or some small domain-specific language. There still are some great hackers writing Java code, just as there are using C and C++, but we are already on the descending part of the curve.

Another point where I think Paul is straight on the money is in telling companies that they should completely rethink the employer-employee relationship.

Without going to such extremes as transforming all your employees into entrepreneurs and funding them instead of paying them a salary, the first thing they should do is provide employers with the very best working environment. If they can’t afford offices with doors, Aeron chairs and 30″ Apple Cinema displays for everyone, at least let them work from home.

The second thing they should do is let developers work on their own projects for some of the time. That’s not enough, actually. They should force developers to devote a consistent amount of time (like Google’s 20%) to side projects. And they could get better results by having the very best people work on tooling rather than fulfilling customer requirements.

Particularly if your business model is based on body rental, you should radically rethink it. There is only one working environment that is worse than working in your own company office: It is working at a customer’s office.

It’s not that you cannot thrive and survive with such a business model. There are lots of companies getting away or even doing very well with it. It’s just that you’ll never be able to hire and retain great hackers if this is how you work.

This kind of advice should be particularly heeded by those companies who pretend to be adopting and promoting the Open Source way of developing software, or are asking their developers to interact with Open Source communities and participate in Open Source projects.

Certain kinds of business models and working environments are fundamentally incompatible with the Open Source way of working. If you think you can have both, you’re deluding yourself. And there’s no doubt about which environment the best hackers would choose.

I am glad to report that the company I currently work for is not totally clueless with respect to these issues. Even though it takes courage to introduce radically new working policies and business models, particularly when you have families to cater for, I am confident that we are moving in the right direction.

Sometimes even catholics “get it”

Vatican wants to end battle with science - Science - MSNBC.com: “A Vatican cardinal said Thursday that the faithful should listen to what secular modern science has to offer, warning that religion risks turning into ‘fundamentalism’ if it ignores scientific reason.”

This is a nice endorsement of Evolution — and the scientific method in general, if you disregard the nonsense about “proof” — coming from a high ranking catholic priest. Maybe we can put to rest our fears of the Catholic Church turning to creationism, after all.

Zvents

Zvents seems to be generating a lot of buzz lately. Scoble even thinks this is a signal of the incoming doom for Yahoo! and MS mapping platforms, and that Google will be the winner, in the end.

Personally, I started playing with it just to see whether it would be useful to link OSZone Conferences pages to Zvents events. The problem, at the moment, is that the events available on Zvents are quite few and limited to the San Francisco Bay Area. You can create events for other locations, but you can’t search by location name then.

So I limited myself to add a couple of Zvents links here and here. It would be cool to be able to automatically perform a search using Zvents APIs when a new conference is displayed and, if a match is found, link to the relevant Zvents page. This however presumes that the conference name as stored in OSZone and Zvents are similar enough to find one and only one match when searching.

Also, I’m still trying to understand what benefit an OSZone user would derive from clicking through Zvents, apart from a cool map. If it’s just the map, we could simply integrate Google Maps directly in OSZone. Now that I think about it, why not?

Open Source Conferences Calendar

CalendarHere’s a new feature for The Open Source Zone: A nice monthly calendar for Open Source conferences.

Head over to The Open Source Zone to check it out. And if your favorite conference, convention or symposium isn’t there yet, feel free to add it yourself.

Of course, you can always use the Conferences RSS feed to stay up to date on new conference announcements, just like before.