1. have two female characters
2. who talk to each other
3. about something other than a man

They may sound obvious, but popular movies fail the test on a regular basis.

The same applies to communities led by a figurehead, such as blogs or fan pages.

Stage Zero happens when there are no participants. It’s a sad place.

Stage One happens when there are some participants who leave comments and interact with the figurehead, but mostly ignore each other.

Stage Two happens when the participants start interacting.

Stage Three happens when the participants, still interacting with each other, start going off-topic and discuss things beyond the original purpose of the community.

Most communities get to Stage One. Yet, even as they get hundreds of participants, getting them to interact together is harder. Hundreds of people go write “Great Post! I completely agree with you!” in the comments section of every article on 10k-subscriber blogs because this gets them some free back links. It takes a lot of spine and insight to actually write something original, contradict the author on his own blog or *shudder* go post on a brand new blog with only a few subscribers.

This blog is still in Stage One. I think Stage Two is a nice place to be, and my regular posters seem to be clever and decisive enough to go beyond the “Great Post! I completely agree with you!” wall. Stage Three would be even better

I had a jolly good time. The trip back to Paris took about 5 hours by train. I wound up back at home a short while after midnight and, since there was no Internet connection available for the entire week-end, I went straight for my Internet fix. Which I couldn’t find, because my CastleNet cable modem had died during the weekend. I scratched at the walls, screamed in the middle of the night, and generally behaved like a junkie for a few seconds.

Today, I replaced the cable modem with a new one for free (Numericable, my cable provider, basically just rents me the modem and replaces it whenever it dies). I’m very unhappy with the new modem because it’s a Netgear cable modem. Netgear.

I’ve had some experience with Netgear routers before. A bad experience. When I was still a teenager, my family moved into a new home and decided to go wireless, so we bought a Netgear router and a Netgear PCI wireless card for every computer. It sucked. First, there were random disconnects as the router would just shut down and reboot of its own accord if you did so much as stare at it. Then, the signal power meant my bandwidth would have been greater if I had just ran around with my packets written down on a piece of paper. And one day, the router just decided life was not worth living anymore. With my last strands of sanity I purchased a Linksys WRT64GL and swore never to buy a Netgear product again. Right now, I’m sitting face-to-face with my latest Netgear modem-router (it’s wedged between my two LCD monitors) with contempt in my eyes.

The Point (yes, there’s one)

So, I judged an entire product line, nay, an entire company based on my experience with a single product that might have been, for all I know, a honest malfunction. Everyone does. You don’t eat a spoonful of Yogurt, decide that it tastes like the zombie apocalypse just happened in your mouth, and buy another pack just in case. I’ve already explained why creating a great first impression is important:

When you manage to attract a truckload of fresh eyeballs, make sure your secret weapon is ready. Show them the fireworks. Have them go Oooh! and Aaah! Every time you attract someone to your website or product, you implicitly promise them that you have something very interesting to show them.

But sometimes, something goes wrong, and the only impression someone ever gets of your brand is a bad impression. What do you do, then?

The common reaction among businesses seems to be «ignore the issue completely» and there’s a fairly simple reasoning behind that: if a one-in-thousands occurrence creates a bad experience for your users and you say nothing, then those few people will be unhappy and stop buying your products, but the vast majority will remain blissfully unaware that there were problems in the first place. This was a fairly good reaction ten years ago. Not anymore.

A one-in-thousands fiasco has a significant probability of ending up on Twitter or on a blog. From there, it’s only a matter of time before that blog ends up with a high Google rank for «[Your Product Name] sucks» and every person looking to buy your product ends up reading it. Seriously, just Google «[Any Product Name] sucks» to find the arguments against any product you can think of. And there’s worse: what if the media finds out about and investigates?

Seriously. Apple tried to hide, and then deny having the iPhone 4 antenna problem, and finally lied that everyone else had the same issues. What is probably the best PR machine on earth after the White House just failed. Can your PR department handle something like this?

What if Netgear themselves had announced that one of their WGRxxx router lines had unusually annoying issues due to a faulty design? Sometimes, all it takes is to sacrifice a lame product line that everyone knew was bad (but no one dared say it) to invigorate the brand.

Oh, by the way: I know today’s post is late. My Internet connection was out.

What do you think about Netgear products (or Linksys ones) ? Have you given up on any brands because of your early experience? Do you have any error-detection techniques that let you target victims of your product failures with an apologetic e-mail?

Related Failures

• 1. Find Crowd – 2. ??? – 3. Profit!
• The Forces of Evil
• Bundles : Not Limited to Products Anymore

• Removed the calendar from the sidebar.
• Added a “Like” button to every post.
• Added a fan page badge to the sidebar, using the newly created Nicollet.net fan page :

Dear readers and subscribers : I know that you’re more than an user agent and IP address! If you like what you’re reading on this blog, please consider joining the facebook page: it’s a simple way to help others discover this blog, you will be kept up to date with frequent updates, and I will get to see your happy little faces

Anyway, below is a simple tutorial for adding Facebook capabilities to a website or blog, just in case.

Adding the “Like” button

The idea behind the “Like” button is that Facebook will keep track of who liked what page on the internet, based on the page’s address. By default, it shows how many people clicked the “like” button. If the friends of a visitor liked the post, these friends will be explicitly named. If nobody has liked the page, it cleverly displays a “be the first of your friends to like this page“. Oh, and you can choose to rename “like” to “recommend” if you think the former sounds stupid.

This button is implemented as an iframe. You can generate the HTML for a certain page by going here and entering the address of the page. You will end up with HTML that looks like this:

<iframe src="http://www.facebook.com/plugins/like.php
?href=http%3A%2F%2Fwww.nicollet.net%2F&amp;layout=sta
ndard&amp;show_faces=true&amp;width=450&amp;action=li
ke&amp;colorscheme=light&amp;height=80" scrolling="no
" frameborder="0" style="border:none; overflow:hidden
; width:450px; height:80px;" allowTransparency="true"
></iframe>

Hidden in this mess is the address of the page to be liked, which means you can generate it using your server. For instance, to get this nice button below every post, I edited my WordPress blog template to add this iframe at the bottom of every post, then replaced the URL address with some PHP code:

<?php echo urlencode(the_permalink()) ?>

It’s important to keep in mind that Facebook determines the number of “likes” based on the URL. So, make sure all “like” buttons for a given page use the same address (for instance, adding a “like” button to forum threads would involve using the link to the first page of the thread on all pages of the thread).

Creating a Facebook Page

This one is exceedingly simple. Go to the page creation form, fill in any required information, and go! If creating a professional page, you might want to create a Facebook account that is distinct from your private account, to use as a page administrator profile. Once the page is created, add a picture, fill in the info tab, and write a short text in the left sidebar. Then, start promoting the page.

Adding the Page Badge to your site

Actually, it’s called a “like box”, and again, there’s a clean creation form available. Fill in any required details, and copy-paste the generated HTML back on your site. The only difficulty here is figuring out your page number. It’s hidden in the URL of your page: for instance, you can find the Nicollet.Net page at

http://www.facebook.com/home.php?#!/pages/NicolletNet/133220253364964

The page number appears at the end of that address, so just copy-paste that value.

Stop Micromanagement. Take any game where you play as the here and have to accomplish something. Now, turn it into a game where you control the world to help the computer-controller hero accomplish the same thing. This is the difference between doing it yourself and micro-managing someone to do it.

Programming Games. On the topic of having games teach interesting concepts, the second installment of LightBot is now available on Armor Games ; the game teaches recursion, recursion-based loops, and conditionals.

Google Street Smarts. It’s fairly easy for us to search online for our own names, to see what others might find. How many of you have tried to search for pictures on geographical locations based on their names? And how can we know if we’re not on Google Street View, anyway?

New Favicon. There’s a new favicon on the blog. You should see it in the address or tab bars above, or here:

The blog originally started back in 2008 on a Joomla! platform on my shared hosting. The first post, Options and References, was published on August 8th and compared the C++ and Objective Caml approaches to option and reference semantics. I managed to follow a thrice-weekly schedule for a few months, forcing myself to write when I felt tired, which resulted in posts that were sometimes badly formatted or completely uninteresting. There’s some serious chaff in those archives, folks.

I started a blog because I was at the time a very active participant on the gamedev.net forums, with several thousand posts under my belt, and many of these were long, in-depth analyses of technical topics. This also helped me improve my english language skills—I’m not a native speaker—and provides me with an invaluable argument in any hiring process. Not bad.

The game development forums were also the reason the blog was written in english : most potential readers at the time were people on the forums, so I had to cater to that audience. It ended up reaching far more people than that, as the blog gradually shifted from a C++/game focus to PHP/JavaScript (and now, to business strategy).

Several blog posts actually come directly from these forums, copy-pasted and quickly edited, and many others were directly inspired from there:

• Shoot’em up Game Design
• Member and Non-Member Functions
• Can references be null?
• An Introduction to C++ Pointers
• Contraceptive-Oriented Design
• Segregation is Good
• Array iterators
• Harbringer of Spring postmortem
• Bored CSS

Actually, my 100th post on the blog (it was published on April 1, 2009) was titled The C++ hits you! You feel confused. and discussed a common topic found on the forums : should C++ be used as a language to learn game development?

Then, I moved to wordpress in late december 2008. I have had nothing to complain about ever since (well, except comment spam). I have been using K2 as a theme, and modified it last month to improve the readability of the blog, to add a Retweet button to all my posts, and to append a small 80-day plot of my feed subscribers at the bottom of the right column.

The Future

I certainly intend to keep posting here. There has been a gradual increase in readership over these two years, which might be related to my own improvements in both content and style. Besides, there are benefits to blogging (and, in general, to writing down things) : it serves as a log, as a way to think about stuff in greater detail, and as a reassuring routine that just stays there when you change jobs or locations.

I also intend to make a few improvements. These would be:

• Add more links. I believe this is the single greatest offender as far as content is concerned. There’s absolutely no way for an average part-time blogger like me to produce as much fresh content as the rest of the internet. This isn’t a plan to become an echo chamber, but rather a way to add depth to my existing posts. There’s a middle ground to be found between the echo chamber blog and the listen-to-me-I’m-interesting blog.
• Add more internal links. Now that I have two hundred posts, there’s absolutely no way for the average reader to skim through the archives to find something of interest. There are so many subjects here that a PHP developer will not care about C++ pointers and a start-up founder will not care about CSS. A short related links section at the bottom of every post should help with this whenever possible.
• Add more images and structure. Right now, my posts are a neverending wall of text and pain. Adding a few more images, headers and lists back into the fray should help make my posts friendlier.
• Combine several unrelated short posts into one. Right now, I don’t publish an article if it’s shorter than 1000 words (or at least, not without painful mental duress), but I have lots of ideas that I could write two or three sentences about. By combining them into longer posts, like Inc.com does (example here), I get to write more, and discuss a larger variety of topics.

On the other hand, there will still be no twittering (I have neither the time nor the interest to take care of that).

I hope the improved version will be to your liking. Until then, if you have any advice (or constructive criticism about existing content and practice), feel free to mention it in the comments section!

• Decreased the size of the header, which used valuable screen real estate, from 220 pixels to 70 pixels. In terms of screen size, it decreased from 1/3rd of the screen height down to 1/9th.
• Increased the width from 750 to 960 pixels (all of that increase went into making the content column wider).
• Increased the font size by a few percent, and reduced the line height a bit.
• Restored sane colors for the links (blue by default, orange when visited).
• Corrected some silly rules (such as strong text using a different font, code being larger than the rest of the line, and the extreme line height in pre blocks).

Hopefully, it should be easier to read.

CT — October 5, 2009 at 22:15

Interesting stuff. I don’t relish the idea of taking the vile HTML our designers produce and creating the skin files. Nice proof of concept though – I’ll have to keep an eye out for an excuse to use it ; )

This comment, while completely adequate and relevant to the article, is spam. How do I know? First, the provided website is a classic credit-rating-improvement web portal. But should I prevent people who work in the credit spam industry from posting relevant comments on my articles? Well, there are other comments on that article, too, such as:

Tom Milsom — September 8, 2009 at 11:41

Interesting stuff. I don’t relish the idea of taking the vile HTML our designers produce and creating the skin files. Nice proof of concept though – I’ll have to keep an eye out for an excuse to use it ; )

So, it looks like the spam-bot found an earlier comment on the article, copied it verbatim, and posted it with a different link. This would ensure that, if the spam domain is fresh enough not to register as such, the Akismet spam detector would let the comment go through unscathed based on its content alone. And as a human, if I did not pay attention to the author’s website while reviewing comments, I would let it go through as well because the comment would look sane. I don’t remember comments from one month ago, and I guess many people don’t.

Everyone enjoys advertising if they are looking for, or otherwise interested in, the product being advertised. I discovered Cushy CMS because it ran an ad on The Daily WTF, and I am quite happy with the discovery because I was looking for such a product. And nobody enjoys advertising for products they don’t need—I don’t give a cheese about US credit ratings. I have limited space on my screen that I’d rather not fill up with advertising about things I do not need, and my time is even more precious than that.

This spam comment blurs the line between spam comments that are irrelevant to the discussion and point to websites irrelevant to the readers, and ham comments that are relevant to the discussion and point to websites that are relevant to the readers (by virtue of usually being run by the author of the comment and thus sharing at least some elements).

Suppose that tommorrow, someone posts an original and interesting comment on one of my articles, yet links it to a credit rating website. Should I accept the comment as such, block it, or publish it without the link?

One of the main reasons why people comment on the blogs of other people is to improve their visibility on the internet. If I post a comment on a well-known blog, hundreds and thousands of people will browse over that comment, a small percentage of these will find my writing worthy enough to follow the link and end up on my blog, and an even smaller percentage will become regulars, posting comments and subscribing to my feeds. Which is good, of course, because the more comments I get on my blog, the more interesting it becomes.

This means that commenting is often quite similar to advertising one’s own blog or website. People allow commercial advertising on their blogs (ad banners and such) to get money in return, and they allow personal blog/website advertising on their blogs to get comments in return. So, I guess if an irrelevant website was linked to by a genuinely interesting comment, I would publish that comment (of course, restrictions do apply: I would not allow all websites, just like I would not allow all ad banners).

I like the blogs with good comment advertising—where I can browse the comments and find links to interesting websites.

• Some CSS page layout templates.
• The classic Silk icons from FamFamFam.
• Color hunter, creating color schemes from images.

• I don’t have an excellent idea for a column, so I’ll write a short article on an idea that’s good, but not awesome.
• I can’t stay passionate about a topic for weeks, nor can I write everything in advance and still manage to fill the less passionate columns in time, which means a lot of ideas are just thrown away because they don’t fit my publishing schedule.

Another reason why I’m giving up on these is that WordPress seems to have wiped away all my categories. There are no more categories either on the public website or in the back-office, and I won’t be hunting for them in the code or database anytime soon.

Also, I won’t be changing the regular schedule: Tuesday, Wednesday and Friday remain fixed publishing days, although if I have too many things to say, I will add them anywhere during the week. This way I still have a quantity constraint (or else I won’t get anything done) and the fixed days will prevent me from writing the articles in a single sunday evening rush.