There will be no Linux Vista by ZDNet's Dana Blankenhorn — Imagine if, in the proprietary era, Microsoft launched a major Windows upgrade while its main rival stood still. That would be advantage Microsoft. Under the present system it may well be advantage Linux.

One of the advantages of open source is that projects release when it is appropriate and not for the sake or releasing or to make more money. The Linux OS is stable and useful, it shouldn't be releasing nearly as often as a project higher up the stack.

Vista's release may prove to be a huge step backwards for Microsoft and hopefully a wakeup call. When people look at Vista they are going to either say "it isn't worth it" or "why should I pay more an upgrade when I can move to Linux?". Either way, I expect it to be a huge problem for Microsoft, especially in the server arena.

I sometimes get asked by aspiring manages and leaders how to deal with firing someone. It is scary to them to think of causing that much pain and that much disruption in someone’s life, as it should be.

Fortunately in the technical world, it doesn’t need to happen very often. In my 25 year (ouch) career I have only had to actually fire three people (discounting the company-wide layoffs in the dot-bomb years). Most of the time, if you are not happy with an employee, then that employee is not happy in the job either. If there is enough disconnect that this isn’t true, then you aren’t doing your job. In most cases, talking candidly with the employee and coming from a perspective of trying to help him or her will suffice. If the person really isn’t right for the position, then these talks usually end in the person getting a different job. Sometimes this is inside the company and sometimes outside; in the startup world, it is almost always with a different company.

When I have had to fire employees, it hasn’t ever been a surprise to that employee – again, if they aren’t well aware of the situation, then you haven’t been doing your job in communicating with the employee(s) enough. The only times I’ve actually fired are when some event happens with an already dissatisfied employee who I’m working with and that event is of a severe enough magnitude that employment can no longer be tolerated. One example of this was a network administrator I had one time who was having difficulties and then bragged to another employee that he had been reading the emails of the CEO; of course this is a serious security violation and had to be dealt with.

As long as you go about it from the standpoint of doing what is right and helping that person as much as possible, then it isn’t as difficult as you might think. Don’t get me wrong – it is still difficult and I stress about it just like everyone else (you would have to be a cold person indeed not to). But, if you are upfront and factual with the employee and you are trying to do what is right for the company and for the individual, then any reasonable person will come to appreciate that, even if he/she doesn’t like what is happening. And really, three times in 25 years isn’t very often.

I finally found a humanist organization I want to participate in. Too many humanist organizations I have seen are radicals; they spend all their time complaining about the current politics and religions and not enough time promoting humanism.

Humanist organizations need to concentrate on making humanism more visible and more viable to the general population. This should include things like public speaking, books, articles, etc. as well as promoting ideas that replace some of the valuable aspects of religion (rites, traditions, social interaction, etc.)

The Boulder International Humanist Institute seems to be much more geared to teaching and evangelizing humanism than whining. They have a great speaker series (recently they hosted Sam Harris, author of “The End of Faith”, which was a great success).

I met with the founder (Gordon Gamm) last week and was very impressed.

Of course their web site is really crude, but that is something I can help with…

Sean

Dana Blankenhorn posted on ZD Net (http://blogs.zdnet.com/open-source/?p=780) about open source's impact on programming in America and about Ludd vs. open source. This seems pretty clear to me; I’m a huge believer in open systems, capitalism and a balanced and fair playing field. If open source puts programmers in foreign countries on an even [competitive], then great! The world is becoming flat (see earlier post “The World is Flat”) and that is a really good thing. It brings us closer together, ties the various countries and makes us interdependent. It also significantly raises the standard of living of people in countries that have previously had few ways to do so. What we get out of all this in the end is the best programmers in the world creating the software for all of us to use. We end up with higher quality software at a much lower cost and the efficiencies of every company (that has the intelligence and foresight to use it) more efficient. The only people who might suffer are the feeders at the bottom of the programming pool here in America. They will be squeezed out by higher quality work elsewhere. You could argue that the lower cost of labor, even professional labor like this, overseas means companies will use lower quality work there rather than the higher paid engineers here; but I don’t really believe that. What company or person do you know of that would say they would accept lower quality (in software) to get a better price. We have an incredible pool here in America of highly creative, talented and capable software engineers. The difficulties and additional costs of sending work overseas makes it difficult to match the productivity of a creative team of highly talented software engineers here in the US. But, we also have a fair bit of “noise” here in the US – people who are software engineers not because of talent, capability, or love, but because of the money. Those are the ones that will lose out here because that kind of [lack of] talent can indeed be found overseas for a lot less money; and good riddance to them; they should go find a field that the love or where they have a natural talent. I have done and do a lot of out sourcing to India and other countries and they have a great pool of very intelligent and well-trained people there (as long as you pick the right companies to work with). But, I still keep the creative, fast moving software development here in the US. This tactic of using each country for the things the people excel at is what will accelerate companies and make them more efficient – and is exactly the way a capitalistic country should operate.

I read Brad Feld's blog periodically and he has a series going about how to run and participate in a board of directors. His most recent post was on how often board meetings should happen. Of course that varies based on the where the company is at in it's evolution. But, at an early stage he suggests monthly in-person meetings. From experience I can tell you that monthly meetings seem like a lot of work and I know there is a lot of time put into preparing for them every month (sometimes it feels like roughly a week every month we are preparing for them since we really start concentrated thinking and effort about a week before the meeting). But, I also know that monthly meetings keep us on our toes. Sometimes we get to a week prior to the meeting and wonder how much we could possible have to talk about since it has only been three weeks since the last meeting; then we start talking and thinking about it and almost always have loads to say and can show how much progress we have made. So, even though it seems like a pain sometimes, having those monthly board meetings is a good thing. First we get very quick feedback and input on our direction and what we are thinking. But also, it forces us to concentrate on progress and it shows us just how much progress we are making. It use to be we had weekly status meetings in the software engineering world; those have moved to daily stand-ups in the agile development world. The same is true at the management level; we use to have quarterly board meetings, but in the more agile world we have to live in, monthly meetings are more appropriate. Of course as a company grows and matures, those meetings should get less frequent and can sometimes be done remotely.

As of this week Pluto is just a dwarf and our solar system, according to the astronomical union, only has eight planets. I know many people are upset about the decision, but really it was the right decision. Anything else would have just been confusing and would have continued the controversy. Personally I think the union made the right decision and I'm sure it was a difficult one given the emotions flying around about the topic. Sean

I was listening to a presentation the other day about open source and the advantages and was surprised to not hear the thing I consider one of the most important. The presenter seemed to me much more financially oriented as he centered around the cost savings in using open source. He also listed a number of other advantages, ones we have frequently heard. What he didn't list as an advantage was the inherent high quality in open source. Imagine for a minute that you are a developer and are going to contribute some code to an open source project. The entire time you are developing that code you know that anyone in the world could end up looking at it and you know for sure that there will be a lot of eyes on your work. Any person in that situation is going to develop the absolutely best code s/he can. Then, if anything is wrong with that code, the developer is much more motivated to fix it ASAP as it is their reputation on the line. I have worked at many different software companies, both developing software and managing groups of developers and I've worked in IT doing similar work. In every one of those cases the checked in code was nowhere near the quality that open source projects maintain constantly. Even with peer reviews and agile methodologies, you still don't have the care to produce the perfect snippet that you would when working on open source. This is one of the big reasons open source projects tend to be so much better than commercial products (of course there are exceptions and this doesn't really count the bottom of the open source pool where it is just one guy submitting code to SourceForge). And this, to me, is one of the biggest advantages of open source – the high quality of the projects we get just because of the model.

I've been at startups for my entire career and at the executive level for most of it. One of the nice things in the startup world is that the dress code has always been fairly slack.

But this seems to be changing. Recently I had the misfortune of having to find a new job. One thing that happened during that process was that it became clear that the interviewers were expecting nicer dress than I was use to. One even had concerns about my ability to present to the board – that made me laugh; I think the concern though came from the fact I wasn't wearing a nice suit and he was. I also go to networking events and I've seen the trend moving more and more back to wearing suits or at least a sport coat. A few years ago at similar events I would see maybe 20% of the crowd wearing ties and the ones I've been to this year it has been more like 70%.

I have mixed feelings about the trend, but I do wonder what is driving it…

As VP of Engineering, I frequently need to translate between engineers and management and one of the areas that come up frequently is our use of SCRUM (agile development). We happen to be lucky enough to have a CEO with a technical past and one that understands development methodologies. Of course this is good and bad in that he knows enough to question me :(

He recently commented that which methodology we use didn't really matter, what mattered was how it was managed. It is true that any of the methodologies can work given the right management. But, there are a number things that agile in general brings to the table that are more difficult (though not impossible) to handle with other methodologies, especially waterfall.

The first is that you should know a lot sooner if things are not working well in an agile environment. In a waterfall methodology, it may be months or longer before you realize you are way off track on a project. In Agile, you should see the "velocity" fairly quickly and be able to point out the discrepancies and adjust.

The next is that you should be able to adjust to changing requirements and changing environment/market much more quickly with an agile environment. With waterfall, you may take months just to nail down the requirements and by the time you are done, they are out of date. In agile, you adjust each iteration or sprint (usually 2 to 4 weeks long) and so are keeping up with any changes as you go.

Then there is the general philosophy in agile that attempts to build a creative, high performance team. This is certainly possible in other methodologies as well, but the framework defined by agile methodologies helps create the right, open environment where waterfall tends to push processes and control.

So… yes, they can each work given good management, but how many really good technical managers/leaders are there out there and I would contend that agile development methodologies help produce better results even if (especially if?) you don't have that kind of great management (or maybe it helps to train the manager :)

Brad's idea here is really interesting:

My Virtual Library

My first thought was what a great thing to do; then I wondered if the recipients would think you were trying to send them a message. I think that could be handled with a short note and it is a really wonderful way to let people know you are thinking of them and of doing something useful with the books.