5 ms·

I think there's a legitimate question of how possible it is to build a stable software/internet business for the long term It's easy to start such a business;

by 13ren 18y ago

I think there's a legitimate question of how possible it is to build a stable software/internet business for the long term It's easy to start such a business; but it's also easy for your new competitor, and for open source. But apart from barriers to entry, there's that routine cataclysm of a computer technology revolution every decade or so. That one is hard to survive. This volatility is the main reason Warren Buffett doesn't invest in computer technology companies. I think the long-term is a decade or so.

garethm 18y ago
For what it's worth, MiTek Australia, a subsidiary of Berkshire Hathaway, recently acquired a software company: http://www.buildsoft.com.au/Content_Common/pg-Announcements-1.seo http://www.buildsoft.com.au/Content_Common/pg-Announcements-...
- 13ren 18y ago
  Thanks for that. I think there's some kind of a distinction between a technology company, and technology company that is directly tied to a non-technology business, so that it also is a non-technology company. Buildsoft (the acquired company) develop software for the building industry. At some point, this is the same business as developing physical tools for the building industry. Once there's a known task and a standardized way to solve it, we can develop our tool for doing that. There's less uncertainty. I think the distinction is to do with uncertainty, generality and abstractness. A "technology" company is doing something new and unknown, that is applicable to many different industries - but hasn't settled down yet. Once the application becomes concrete, and specific to a particular industry, and the way of solving it is pretty well established, I guess it becomes just a regular business like any other. Moore's "Crossing the Chasm" talks about customizing a product to a specific industry, as a way to get a measure of security - you won't get wiped out (until the next revolution in computers). I was thinking that Warren might not have anything to do with this acquisition, but he's very attentive to capital redeployment, so I would guess he did personally approve this purchase.

by 13ren 18y ago

garethm 18y ago

For what it's worth, MiTek Australia, a subsidiary of Berkshire Hathaway, recently acquired a software company: http://www.buildsoft.com.au/Content_Common/pg-Announcements-1.seo http://www.buildsoft.com.au/Content_Common/pg-Announcements-...

13ren 18y ago
Thanks for that. I think there's some kind of a distinction between a technology company, and technology company that is directly tied to a non-technology business, so that it also is a non-technology company. Buildsoft (the acquired company) develop software for the building industry. At some point, this is the same business as developing physical tools for the building industry. Once there's a known task and a standardized way to solve it, we can develop our tool for doing that. There's less uncertainty. I think the distinction is to do with uncertainty, generality and abstractness. A "technology" company is doing something new and unknown, that is applicable to many different industries - but hasn't settled down yet. Once the application becomes concrete, and specific to a particular industry, and the way of solving it is pretty well established, I guess it becomes just a regular business like any other. Moore's "Crossing the Chasm" talks about customizing a product to a specific industry, as a way to get a measure of security - you won't get wiped out (until the next revolution in computers). I was thinking that Warren might not have anything to do with this acquisition, but he's very attentive to capital redeployment, so I would guess he did personally approve this purchase.