Open Source Coming of Age

I found this perspective very interesting. It’s called “Lack of testing ‘threatening stability of Linux.’”

We have been seeing a lot of things like this, where people identify that how we do open source needs some re-thinking. There have been others declaring that we need to deal with the association of open source to “free” (in the sense of “free beer”). As open source comes of age and gets put to increasingly higher-end use, we can expect a lot more of this kind of dialog. It’s healthy. We need to embrace it.

LinuxWorld’s “Why Is Open Source World Domination Taking So Darn Long?”

Asking the Wrong Question can Make your Point
I recently ran across an article on LinuxWorld.com entitled “Why Is Open Source World Domination Taking So Darn Long?” Christian Einfeldt makes the right point about the delayed domination, and he does it with a red herring that started me off on a very different rant. But, it turns out that the questioning title of the article actually answers itself. That is, there are enough reasons for it taking “so long” that the act of asking the question demonstrates how so many of mislead ourselves into overestimating the general “readiness” state of open source software.

How We Fool Ourselves
Many of us look at the ideals and vision that open source sets–collaborative development and shared success, a certain cooperative altruistim, and better software through some sort of benign Darwinian progression–and we see the future now. The process is better. The motivation behind producing open source software have a pleasant infusion of altruism. The quality and security of the code through the open source development process is generally better. The intellectual freedom it repesents is inspiring. All around, we can see a convergence of evidence to indicate that open source software is just better. It comes from a vision that is compellingly superior to how software has previously been produced. But there’s one problem.

Many of the products of open source have yet to surpass their proprietary software counterparts. For that matter, a lot of it isn’t even “good enough” yet.

To be sure, it’s not always because of the actual software. Often it’s because we are so quick to presume the “for what?” that necessarily must follow “good enough.” (That, in fact, is one of my ongoing rants.) The open source idealist often jumps past many of the reasonable expectations that end users have for software usability. Enthusiasm for open source is good to have, but taking it too far (overzealousness) leads to setbacks.

Even though it won’t happen overnight, three things will lead to the eventual dominance of open source. First, finding right-fit roles for various available open source projects. Second, learning from their use in those roles, and the gradual progression of the projects as a result. Third, not overshooting or promising beyond the capabilities of the software at any given time, because each failure can become an example–large or small–of how the software is “not ready yet.”

The general “it’s not ready yet” stigma is the worst thing for open source, and we the believers in it only harm our own cause when we force that perception by overshooting.

Conservative Positioning
This need for even-handedness with open source is why Novell has received some amount of inquisitive commentary about being too conservative in positioning Novell Linux Desktop. (Even some of my Novell colleagues have said so much to me.) My team and I stick by a primary message that Linux desktops are best when applied for specific tasks and well-understood user types. (Fortunately, our CEO is of the same opinion: bullish on the desktop opportunity, but realistic about the products best-fit roles within an organization.)

For the record, I compose this blog from my Linux laptop, usually using a wireless connection that I can easily switch to at home or the office by using the desktop wireless applet (netapplet) that was included in Novell Linux Desktop. Sometimes I compose in OpenOffice.org Writer, but usually I use my Firefox browser directly. So, I for one am comfortable working from Linux, and I am one of these “knowledge workers” that I tell most organizations to target last. I rely on Linux as my primary productivity operating system. The software is indeed “good enough” for me.

But here’s the rub: my wife–a marketing professional as well–is still not comforatable in Linux. She’s savvy enough with a computer to feel the limitations that OpenOffice, GNOME and a host of not-yet-available applications put on her. Since it’s not right for everyone yet, I stick to the idea that if we’re up front about the technology, rather than overcome with our own optimism, then new adopters (particularly IT decision makers) will approach it with the right mindset when they start looking into adopting Linux desktops.

Performing as Expected
Back to Einfeldt’s article…Open source world domination is following exactly the course one would expect an emerging and disruptive technology to follow. Most new technologies don’t take over like wildfire, suddenly bursting into flames of general popularity. Instead, new technologies usually have to get grab in a few select spaces and then work their way to general popularity. This is particularly true with replacement technologies. (When I travel abroad, I am often surprised to find that the wireless telephone infrastructure in many developing countries is actually ahead of the United States’.) Newer, often less-expensive technology replaces the older entrenched technology much more slowly than adoption. And Mr. Einfeldt makes that relevant connection–the developing world will be a rapid growth area for open source software, particularly operating systems like Linux.

So, to expand on Mr. Einfeldt’s point, open source software in fact is rapidly growing into a breadth of new roles and adoption is very healthy. People are taking it seriously. But it’s still very early in the game, and it’s likely that there remain several more steps and stages to go through before the revolution is done.

TiVo needs to be Identity-driven.

As an avid TiVo devotee, I’ve been thinking a lot about how TiVo could be made better with some basic identity-enablement. TiVo is a Linux-based personal video recorder (PVR), so I have some interest from the “it’s Linux” part. But having worked at Novell for several years has trained me to always consider how identity could make something better. Here are my thoughts on this.

Two things you could do immediately with some basic identity capabilities in TiVo are:
1. Identity-based program management for families
2. Social networking based on TiVo

1. Identity-based program management for families

TiVo includes some basic parental controls that allow parents to control what content can be recorded on a TiVo. When you look at what’s included, though, one quickly sees that it’s either an unfinished implementation, or the TiVo engineers didn’t quite understand the whole problem that families would face with a personal video recorder. I put emphasis on personal, because it seems fairly apparent that a lot of single/no-children people created TiVo. (I recognize the thought pattern because I spent a lot of time being single, and still have no children.) When it comes right down to it, TiVo was not designed for multiple users. So, way out beyond parental controls on content, an identity infusion for TiVo could be huge for families.

The problem started becoming apparent when I got my first TiVo, which coincided with the early days of my relationship with Kim. Kim, it turned out, liked much different television programming from what I did. That created my first understanding of this problem. (How could Ally McBeal possibly be more important than Nova?!) Then, when my friend Erin got a TiVo, the problem suddenly became acutely poignant. Sponge Bob dominated his TiVo’s “Now Playing” list. (Nickelodeon airs a lot of Sponge Bob.)

So, what would make a lot of sense is to allow a good system of familial prioritization–beyond what the Season Pass Manager currently allows. And, more importantly, disk space management.

If you have kids, it makes sense to me that you would want to constrain how much space the little bastards get to use on your system. Perhaps more important would be the ability to restrict your favorite, recorded programs (which may include a lot of boobies) from your kids being able to view it. TiVo has already implemented folders, which could be quickly adapted to this purpose. Some simple access controls on a folder and we’re there.

The TiVo that we use today was clearly designed by single people with no kids. As a married person sans kids, I can see the gap pretty clearly, as well as the bridge to solve it. (I congratulate myself often on being wise beyond my years. And for my keen sense of potty humor–very good at that.)

The question is really whether TiVo will get on this soon, or whether MythTV, now gaining in popularity among the geek base that TiVo needs to keep appealing to, will get there first.

2. Social networking and TiVo

TiVo’s popularity is predicated on the idea that broadcast television sucks. There’s nothing on. Channel surfing is a huge time waster. And the shows I like are not on when it’s convenient to me. Also, I hate commercials.

Social software like del.icio.us has shown that a social network can form around like-minded people sharing their interest areas through their bookmarked web links. The more you play with del.icio.us, the more you start to see how it can get you to new content upon which you might not have otherwise stumbled. (The only problem is that web browsing requires reading, which is a fairly active mental activity. Watching television is not, and lends itself well to social networking around content.) However, Web sites are not so different from TV programs. We try to watch those that appeal to us, just as we spend time reading sites that interest us.

Here’s the scenario: Recently, I was watching a National Geographic program on human evolution. During the show, it occured to me that I would really like to tell my friend Erin about it. However, TiVo has no feature to allow me to share the program with Erin. And that’s not to say that my TiVo would have to transfer it’s recorded content to Erin’s TiVo. The show will most likely be aired again, so Erin’s TiVo could just record the next showing.

Ideally, Erin would be able to include me in his social network, and authorize me to suggest programs for his TiVo to record. His TiVo could then record “Ted’s Suggestions” much like it already records “TiVo’s Suggestions” as low-priority recordings. And if Erin could perhaps prioritize his different contributors in his trusted TiVo social network, he could ensure that he gets what he considers to be the best programming first.

TiVo of course gets a certain amount of gain from this. The algorithms currently used for selecting programs to record absolutely suck. No further proof is needed beyond the joke from a couple years ago about “my TiVo thinks I’m gay.” The coders at TiVo can offload the “what would he like?” question to the far better fuzzy logic calculators (brains) that reside in a person’s social networks.

I’ve spoken enough. TiVo, make it so.

The Truth about my Cheesy Headshot

I’ve never seen the need for having headshots. On some speaking occasions, I get asked for one, and then there’s a bunch of scrambling to figure out how to provide something.

That’s what happened for the desktop summit headshot that I have been using for this blog. The picture is almost intolerably cheesy, but out of laziness I have left it up.

To come clean about it, it’s actually a really bad photo that Jeff Allen retouched and somehow made look good. This is the original, which shows the truth: I was in an autorickshaw in Bangalore, India. I was uncomfortable and scared out of my wits. Sadly, from that disposition, I can actually look decent. (Said my wife when she saw the retouch: “Who is this beautiful man?” Said she on seeing the original: “Oh, that’s more like it….But, Jeff is really good with PhotoShop, huh?”)

Yes! We Have No Shirts

It is rumored that at Novell’s recent BrainShare conference, Erik Dasque and Miguel de Icaza [please note how Erik got top billing to Miguel...but I'm not bitter] ended their Mono demonstrations [which, btw, rocked] by throwing out some Novell Linux Desktop t-shirts.

Apparently, the looked something like this:

I received the following email request (excerpted) from one Jon Strickland, President, Triangle Novell Users’ Group, pursuant to the Novell Linux Desktop “Decal Kits” we passed out:

I’ve also been trying to find out more details on the NLD 9 decal t-shirt I saw someone wearing in the Solutions Lab. [Ted: that was probably me.] I had been searching Google to see if anyone else had commented on the decal kits or the t-shirt and I ran across your posting on blogger.com. I noticed you mentioned working to get the decal kits offered “through some kind of online ordering system” but are there any plans to make more NLD 9 t-shirts? Our group would be willing to buy some t-shirts if we had a contact to order them through.
…
I guess I’d be classified a “Red Army” member trying to further the Novell cause, and part of that would be helping to build “the Cult of NLD.”

My response to this was:

Jon: Thanks for the interest. As it happens, those shirts are currently valued at about an equal cost as the same weight in cocaine. Coincidentally, under Novell’s current marketing policies, the shirts are considered to be about as legal. So, essentially, you want some contraband. Fortunately, you’ve come to the right guy. In the tradition of the trade, the first one’s free, kid. I’ll dispatch a couple shirts your way, on the q-t. [Etc., etc. --Ted]

Before you flood me with “me, too” requests, allow me to say: that deal was “first come, only served.” But far be it from me to squelch the enthusiasm my company’s most avid proponents. We’re working on a way that we will make these shirts more widely available.