Wednesday, August 06, 2008

Build Times

Oh I wish that the product that I work with could have build times like the linux kernel.
For example, for the kernel, I can literally rebuild my whole kernel (which is just what I use on _that_ machine) in about 16 seconds. This is _not_ using ccache or anything else - it's rebuilding the whole tree with -j16.

Obviously Linus has a machine with at least 8 cores if not a machine that has 4 quad core processors. Also, he has enough memory to keep the entire kernel tree in the buffers. In addition, the kernel is highly modular so there aren't as many dependencies between systems. This means that there aren't a lot of extraneous header files that would slow down the build.

If you know about agile, you know why this is important. Fast build times means:

  • An automated build that can let you know at the minimum that your project is/is not building.

  • Fast builds "probably" means that your product doesn't have a lot of dependencies if it is large and this lends itself to incorporating tests and running those test via automation as well.

The above is the beginning of continuous integration. It can help your software quality especially if you take it seriously and start doing the other things that go along with Agile processes.

Sunday, August 03, 2008

Diets

This is a follow up to the Bias and Diets post earlier. The latest long term study on diets suggests that low carb diets work better than the low fat diets.
The biggest weight loss happened in the first five months of the diet — low-fat and Mediterranean dieters lost about 10 pounds, and low-carbohydrate dieters lost 14 pounds.

By the end of two years, all the dieters had regained some, but not all, of the lost weight. The low-fat dieters showed a net loss of six pounds, and the Mediterranean and low-carbohydrate dieters both lost about 10 pounds.


Another article about this study has more information:
For people with cholesterol problems, the low-carb diet seemed best; for those at risk for diabetes, the Mediterranean diet provided more health benefits.

"The take-home message should be that we should abandon the idea that low fat diets are the number one way for people to lose weight –- it wasn’t the best diet, it can be helpful for some people, but overall I think the first choice should be the Mediterranean or the low carb," he said.

Dunbar's Number, Rule of 150

I came across Dunbar's Number or the Rule of 150 while reading the The Tipping Point. I've noticed that the small companies where I've worked have been less than 150 people. BMH was approaching 150 employee's when it was purchased.

Since being bought out by a much larger company, the BMH operation has changed quite a bit. It no longer has the same feel even though I work with the same basic people. This probably has to do with the way in which the more common processes have been changed and the fact that one of the founders of BMH has become part time as he transitions into retirement.

Don't get me wrong about the above statements. I am not trying to be derogatory toward my current employer. It's more of comment/observation about the fact that small companies that want to grow need to realize that will need to change or they won't be successful.

Management has to prepare for the change and make that change easy or else they will lose the people that made them successful. (They may lose them anyway.) The Hutterite communities mentioned both in the article and the book recognized this early on and have a rule that communities must split once they reach 150 members.

Wednesday, July 30, 2008

Bias and Diets

Here's an interesting blog post about Bias and Diets.
Why is Taubes so interested in bias? For several decades, it has been the conventional wisdom that dietary fat (and especially saturated fat) contributes to obesity, heart disease, and cancer. Judging from Taubes’ exhaustive research -- indeed, I'd be surprised if any other book examined bias within a particular scientific field in such detail -- the conventional wisdom was based on unreliable and slender evidence that, once established and institutionalized in government funding, set a pattern of confirmation bias by which further research was judged (or ignored).

I recommend that you read the article and read the interviews at the bottom. If there's this much bias just for diets, it makes you wonder about the global warming debate.

Sunday, July 13, 2008

Tomato Firmware

After listening to the discussion regarding wireless routers n the TWUUG list, I finally decided to buy a WRT54GL. About 4 years ago I purchased a Linksys WRT54GS v1.1 with the intention of replacing it's bios with one of the custom firmware releases. I never got around to it and I didn't want to brick my only router. So I bought the WRTG54L since it's the only one of the WRT54G/GS models that you can put Linux on now. All of the others have limited memory and only run VxWorks.

I installed OpenWRT. It installed without a problem. I wanted a nice webgui, but the latest didn't have one or I didn't find it. Instead, I used Tomato.

Tomato's webgui is nice. I set it up within a few minutes and replaced the old router with the new. The network statistics are cool. It provides a realtime graph that can be viewed in the browser and it maintains a daily, weekly, and monthly log.

I've turned on QOS, but I don't know how well it is working. This week should give me an idea of how well the QOS works. I set my Vonage voip device to have the highest settings.

I need the extra router for a trip to the beach later this month. After that I'll play around with OpenWRT. There are some interesting things that can run on the router. I want to play around with Asterisk. I've got some actual voip phones that I can use.

Tuesday, July 01, 2008

CC licensed images

Here's an interesting post about the usage of Creative Commons material available on the internet.

Monday, June 30, 2008