You're looking at Less Wrong's discussion board. This includes all posts, including those that haven't been promoted to the front page yet. For more information, see About Less Wrong.

Lumifer comments on Open thread, September 8-14, 2014 - Less Wrong Discussion

5 Post author: polymathwannabe 08 September 2014 12:31PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (295)

You are viewing a single comment's thread.

Comment author: Lumifer 12 September 2014 07:20:04PM 5 points [-]

Y Combinator published a list of requests for startups.

Comment author: Azathoth123 13 September 2014 01:54:09AM 4 points [-]

The list makes for interesting ideas. Most of them seem good, but a few make we wonder about Paul Graham. Some of the ideas (e.g., Government) make me wonder if he's starting to drink his own cool-aid and it has caused him to forget everything he has learned along the way. With others (e.g., Diversity) one almost gets the impression that the SJ crowd is putting the screws on silicon valley and he has to at least through them some bone (the since deleted "Female Founders" essay reads similarly).

Comment author: Lumifer 13 September 2014 03:53:49AM 3 points [-]

I am not terribly impressed by that list as it looks like a collection of wouldn't-it-be-nice-to-have wishes.

The Government section looks fine -- the government is a big customer and does have very bad software. But yeah, the Diversity section is... weird. At least there is no Save the Environment section.

Comment author: Azathoth123 13 September 2014 06:04:46PM *  5 points [-]

The Government section looks fine

It suggests someone at Y Combinator now alieves he has magical superpowers about cutting through government procurement bureaucracy.

Comment author: Lumifer 16 September 2014 02:56:22PM 1 point [-]

Not quite. This is a list of requests -- the Y Combinator would like to find ways to achieve magical superpowers to cut through the government procurement bureaucracy.

Comment author: Azathoth123 17 September 2014 04:17:45AM 2 points [-]

Then why did the section talk about how inefficient government software was rather than cutting through procurement bureaucracy?

Comment author: Lumifer 17 September 2014 02:42:17PM 1 point [-]

Because you need to have what's called a "market opportunity" to start with.

Comment author: Khoth 14 September 2014 07:01:30PM 0 points [-]

A while back I worked at a startup (20-ish people) that had (UK) local government as their main customer.

Large companies don't have a monopoly on providing bad software to governments, even if they have advantages.

Comment author: Pfft 13 September 2014 03:28:56PM 2 points [-]

make me wonder about Paul Graham

I think this list is due to Sam Altman. He has written about wanting to fund breakthrough technologies, and shortly after he became Y Combinator president they invested in a fusion energy company.

Comment author: Azathoth123 13 September 2014 06:06:29PM 4 points [-]

Well, that would explain why the list ignores Paul Graham's advise of investing in fields one understands.

Comment author: Izeinwinter 13 September 2014 10:13:18AM *  -2 points [-]

It seems very credible that "Write better software for the US government" is a field that is shockingly underexploited, simply because of the ideological biases and likely background of the typical american start-up entrepreneur. Do you have the faintest Idea what software to make social services more efficient ought to look like ? Because I don't and I figure very few people looking to start a coding shop do either. The only idea in this field I can think of with any chance of working is to try and run arbitrage against "not invented here" and check what tools are in use in the rest of the first world.

Comment author: Lumifer 13 September 2014 02:36:03PM 5 points [-]

It seems very credible that "Write better software for the US government" is a field that is shockingly underexploited, simply because of the ideological biases and likely background of the typical american start-up entrepreneur.

It seems very credible that this field is "underexploited" for two main reasons.

One is that business dealings with the US government are very stupid, inconvenient, and annoying. You drown in paperwork, you have to certifiy all kinds of silly things, etc. It's OK for a large organization with a compliance department, it's not so good for a startup.

Two is that government contracts are a prime field for crony capitalism. You will be competing not only on price and quality but also on the depth of the old-boy network and the ability to provide invisible kickbacks -- again, not a strength of startups.

Comment author: Izeinwinter 13 September 2014 08:06:29PM -2 points [-]

.. Are you speaking from personal experience of selling things and services to the government here? Because if the answer to that is "No" you may, possibly, want to check if you remembered to remove those ideological blinders I mentioned. The main point of the paperwork vendors to the state have to do is to make sure that crony capitalism doesn't happen. If the process is very badly designed, that fails, but I've never worked anywhere that found it more obnoxious to do business with the government than with any other large customer. Usually it is less so.. The USG can't be that much worse than the nordic countries. It's still a first world state.

Comment author: Azathoth123 14 September 2014 08:39:33PM 3 points [-]

I can speak from personal experience. An executive for the contractor I work for was caught with a massive undisclosed conflict of interests.

This had two main effects: we must now listen to an annual mandatory ethics briefing, in addition to all the briefings and paper work inflicted on us from previous misbehavior. (Note the things talked about during said briefing generally have nothing to do with what the executive was caught doing.) Second, the executive was summarily fired and managed to fall upward into a high level job with the agency we contract with.

Comment author: drethelin 13 September 2014 07:26:36PM 0 points [-]

These are all crazy vague.