Lord of the Tweets

This is an idea I had a while ago, after reading some fluff piece about how Twitter is destroying our brains yadda yadda yadda. Here are some samples of Lord of the Rings, in 140 characters at a time, with some extra pop-culture attitude thrown in.

Bad Hotels (yes, I checked the chronology):

Frodo: Prancing Pony in Bree gets zero stars from me.
Gandalf: @Frodo You should see the place I got stuck at a couple of weeks ago.

Breakup of the Fellowship.

Frodo: Later doodz.
Aragorn: @Frodo Wait!
Samwise is now following Frodo.
Gollum is now following Frodo.
Boromir: Ow.
Pippin: Mmmf.
Orcs: LOL.

Feel free to add your own. I’m sure somebody else can do better, and other books are fair game too.

Inspiration and Perspiration

Edison famously said that genius is 1% inspiration and 99% perspiration. I don’t know about genius, but it’s certainly true in high-tech business. Just about every engineer, sooner or later, has a really good idea or two. It’s the nature of the job, and the people who do it. In many cases, the person thinks about the idea for a while and decides it’s good enough for someone to build a business around, but maybe it’s too big to do in their spare time while holding down a job etc. and they don’t have the resources to develop it on their own. What then? Well, imagine you’re such a person, and that you’ve decided to look for someone to fund development of your Great Idea.

  • You have no technical development plan.
  • You have no market analysis or go-to-market plan.
  • You have no patent protection.
  • You have no discernible record of turning your ideas into actual revenue.
  • You have no significant resources of your own, other than you and your idea, to contribute.

How do you think investors will react? How should they react? Probably with nothing short of derision. SiCortex failed to attract funding despite having all of the above, even having actual built products in real customers’ hands, because investors still saw too much risk. Ditto Woven and Quadrics. But hey, let’s say that you manage to get a hearing anyway. Issues are raised, about technology and about markets. Several times you’re caught by surprise, but you manage to come up with an answer. You probably think you’re pretty nimble, that you’ve shown you can think on your feet, etc. The investors think you’re a fool (or worse) for not doing your own due diligence before expecting them to. They wonder about the issues that they weren’t able to uncover, that you – the supposed expert – should be able to anticipate but probably didn’t because you didn’t spend the time. Then they walk away.

Don’t think it will be any easier seeking a “partnership” with some established company, either. Now you’re not dealing with people who are spending Other People’s Money, and who have an explicit mandate to invest speculatively. Now they’d be spending their company’s own money, pulling it away from other internal needs and thus from other people they already trust and/or feel indebted to. Every standard gets higher, not lower. What you’re basically looking for in this situation is not really an investment or a partnership but a gift. Go talk to Santa Claus.

If every engineer has some great ideas, then great ideas must be pretty common. Things that are common aren’t very highly valued in the market. What distinguishes a true inventor from a dabbler is not having the great idea but being able to have the second and third and hundredth idea that will make the first one work. “Build a better mousetrap…” presumes that you have actually built a mousetrap, not just a part of one and certainly not just slides or fast talk about how mousetraps can be made better. You know that “elevator pitch” everyone’s always on about? That’s not the real pitch. That’s just a way to get permission to make the real pitch, which had better be fleshed out and ready when they call back – i.e. faster than you can Google up all the stuff you should have researched before. It’s all about the ability to execute, and you demonstrate that by having everything “shovel ready” before you ask someone to fund it.

By all means, people should pursue their dreams. I’m not saying for a moment that they shouldn’t. I might even start something myself some day, though probably not this year. All I’m saying is that your dreams can be in the clouds but you have to pursue them on the ground. You have to be realistic, pragmatic, and diligent. Don’t just assume that whatever skills you have are sufficient, and whatever skills you don’t are so low-value that they either don’t matter or you can acquire them at will. Figure out what skills you’ll really need, and either start developing them yourself or forming mutually respectful alliances with those who have them. If your idea is really that great, then it deserves the best support it can get.

War is peace, bondage is freedom, …

From Apprenda’s website.

Rather than requiring the use of a proprietary technology stack, language, or a limiting online WYSIWYG environment, SaaSGrid applications are built using the industry-hardened Microsoft .NET stack and the SaaSGrid SDK.

I guess it’s technically true that using .NET doesn’t lock you into a particular language, but it sounds like SaaSGrid very much locks you into a particular stack. It’s not even clear, with all the talk about Visual Studio integration and such, how many languages are really supported or whether you are in fact locked into Microsoft languages on a Microsoft runtime. Terrible name + blatantly untrue claims = time to get a real marketing person.

DataSlide in, Rock out

In: DataSlide.

1. A piezoelectric actuator keeps the rectangular media in precise motion
2. A diamond solid lubricant coating protects the surfaces for years of worry free service
3. A massively parallel 2D array of magnetic heads reads from or writes to up to 64 embedded
heads at a time

I remember seeing a presentation on something like this at CMU’s Parallel Data Lab several years ago. Very cool. Here’s a fun question: how do you do optimal data layout and request scheduling for something like this, differently than you would for a spinning disk? (Hat tip to StorageMojo for the news.)

Out: Sun’s Rock processor. This is a real shame. I don’t know if I’d agree with all of the directions Rock was going, but it represented some bold choices and I respect that. Maybe the Ice9 team and the Rock team can get together somewhere to shake things up.

Beyond Filesystems

Historically, I’ve been a bit of a filesystem purist. If somebody called something a filesystem that wasn’t fully integrated into the kernel so that any application could transparently read, write and even map files within a familiar hierarchical directory/file structure, or if it didn’t support consistency at least as good as NFS (preferably better) I’d be quick to correct them. I’d say that what they had wasn’t a filesystem, with the clear implication that it was something less than a filesystem. My views have softened a bit over the years, though, and now I see a lot more value than I used to in data stores that aren’t filesystems. I’ve always seen value in databases, of course, but over the last few years more and more people have been putting data into things that are neither filesystems nor databases. In many cases, the value of these other data stores, and/or of what’s stored in them, exceeds that of the filesystems and databases (including the quasi-databases that I’m sure database purists complain about the same way I have about quasi-filesystems). At first I wondered why people were doing things this way. Many of these folks were doing distributed programming. Were they simply ignorant about distributed filesystems? Were they too lazy to figure out the wheel that already exists before they went and invented their own? Those two explanations seemed most likely for a while, but it turns out that there are better reasons as well, and it’s those reasons I’ll explain here.

Job Hunt

As most of the people reading this know, I’m looking for a new job. This is a special post that I’ll be keeping at the top as long as that’s the case. Here are some notes for prospective employers: