pgroves

But why does Greg need a cofounder?

...and is the second investor today to turn on Uber.

Yeah. Does anybody know why walls for houses aren't built in factories (with electrical, plumbing, insulation, etc already inside)?

Health Insurance / Hospital.

I'm old enough that I now go to the doctor more frequently than I used to and it's a mess. A health insurance company that could reliably allow a user to change their address on a website would be competitive. Having all of your medical procedures and orders accessible through a simple CRUD app would be a threat to a lot of multi-billion dollar companies. It's still all done through phone calls and faxes and there are lots of mistakes and it's hugely inefficient. I went to the ER last year and got 5 different bills from different departments of the same hospital. The online payment portal doesn't work unless you call them to set it up. That's not the hospital network I usually go to - my usual provider is probably worse.

I swear this is what Infochimps used to be, but now I don't really see a reference to it on their website. Except for a 404 when I click on "Resources" -> "Data Marketplace"[1]. I'm guessing that means they moved away from that business. Looks like they now focus on tools, not data itself.

[1] http://www.infochimps.com/marketplace

I work at a gov't organization and was going to say the same thing. It recently took me two months to get an $11/month pingdom account. There is some office somewhere that has to read the TOS and put the vendor on an "approved" list. They never use credit cards, and have to call the vendor (even for small webapps) to get bank routing numbers. Also, no recurring payments are allowed, they must be able to pay one lump sum. By the time I get an account the vendor already hates me.

You're maintaining two apps anyway b/c there is the client and the server. The OP didn't say it explicitly, but I'm assuming they are already building the server side no matter what and need to at a frontend. The backend development already builds a bunch of objects that will need to partially reimplemented in the frontend. It's just a matter of how many of them.

My two cents (to the OP) is that you should consider the people or teams building the thing. If you have one team doing all frontend and backend, doing it (almost) all server side makes sense. If you have a dedicated person doing the frontend, commit to a js framework and make a good API to the backend. Certainly don't make a frontend JS guru render html in Java or .NET.

My story is very similar except my insurance company went out of business yesterday:

http://www.chicagotribune.com/business/ct-land-of-lincoln-sh...

This is sort of a tangent, but if you feel you have to deal with too many people to do even a small task, you could make a point of volunteering for larger chunks of work to get out in front of these problems. If the technical people are tripping over each other because of the bureaucracy then make sure every time you have to deal with other people that you are getting a bigger win for you and your team. It would have to be extremely dysfunctional if you said you wanted to do bigger projects because you think you can get more done with the realities of the organization and they said no.

My main two:

1) "Overview and Zoom" is a common UI technique where you give an overview and then let a user click on stuff for more detail. In programming, this means writing the "intention" as a method that's very easy to read, then have a bunch of helper functions with the details. This also forces you to only use one level of abstraction per method.

2) Only have one thing happen per line. This forces you to give what you're doing a proper name (as a variable assignment before you use it). This makes it easier to read, but makes all text processing tools work much better with your code. Nothing is more annoying than seeing a diff in a pull request where one line changes, but there are 10 things happening in that line and only one changed.

"Write Failed: Success" is my new favorite error message.

If there is already a prototype built:

1) What kind of usage numbers "look good"?

2) What kind of ROI on adwords campaigns (or whatever) "look good"?

Or maybe it would be better to frame these as null-hypotheses: E.g. what makes an early adwords campaign look like the whole idea should be scrapped?

I really want this. I gave up on Ocaml because it just isn't a big enough ecosystem but everything else seems like I'm making the wrong things explicit in the code.

Are there ever posts about who has left? Has anyone left?

Seems very similar to how Berkshire Hathaway treats it's companies. CEOs run the companies, a small group at the top adds/removes companies. There is one company listed on the stock exchange, etc.

It's noteworthy that Berkshire Hathaway refuses to deal with technology companies, while Google is exclusively tech.

It seems that way until you find yourself in a situation where none of this advice is being followed. The opposite of doing these things doesn't look terribly sinister, just has poor results. A founder/CEO that pressures the engineers to cut corners on the product to get a few more customers before the next fund-raising round is an easy trap to fall in to. The investors will often be providing this pressure directly. (E.g. "I'd invest if you were showing a little more growth.") It all seems very reasonable when you're in the middle of it.

Another minor UI issue: The Title for the page is "Yelp for..", causing the tab I have it in to have the visible title of just "Yelp" with your logo.

Suggestion: Repeat the the column headings every 20 rows or so or make it sticky. By the end of the page I had to scroll back up to remind myself of some of the columns titles.

If you just want the list of names of the apis:

    curl "http://apis.io/api/apis?limit=1000" -s |python -m json.tool |grep "\"name\"\:"

It sounds like there is enough slack in the schedule that teams can decide they want to spend non-trivial amounts of time on these projects. It's surprising to hear that anyone, even the companies with big budgets, are able to hire enough people to do this without the projects getting an official seal of approval and budget. Even just taking the time to document, package, and publish is non-trivial.

It seems like the vast majority of companies are not far enough out in front of their production issues and requests from the business side that engineers could do this sort of thing. So I guess it's impressive that Facebook (and probably Google) are in that position.

Does someone know how these types of projects come to fruition in a big company like Facebook? Are people working on them full time (with no other workload)? Do engineers build them on the weekend? How do they get 'funded'?

Something similar that didn't make it (shut down in May 2013): http://quant.ly/

I'm actually facing a decision right now in building a new app that has relatively heavy needs for data storage and number crunching. Each paying customer would have a relatively large amount of data per person compared to your average CRUD app.

I'd like to use DigitalOcean or Linode VPS machines for processing and as web servers, but then I need a large object store like S3. Rackspace and AWS have both object storage and VPS, but their VPS machines are underpowered for the price.

So ideally I would use Rackspace-Files or Amazon-S3 for storage, and DigitalOcean for number crunching, but I'd get killed on transfer rates from S3 to DigitalOcean (for example). Amazon and Rackspace have a trump card with free data transfer within their data centers. You have to use their slow machines to realistically use their object stores in this type of a use case.

So that's the 'S3' problem I need solved, which still isn't quite what S3ForMe is going after. Extremely cheap data transfers to/from major VPS providers.

I could minimize some of the fields, but the larger point I should have made is that when I was typing the regex the first time, I was looking at the cheat sheet. The workflow I found myself fall into was:

1. Enter the test string.

2. Look at the cheat sheet and type into the regex field.

3. Look at the matches and test field to see if it was correct.

4. If incorrect, look at the explanation.

5. Go back to (2)

So for me, the default layout didn't handle my step (2) very well. I'm not saying that's the workflow you should optimize the UX for, but it's what mine was.

I like it.

How do you enter a tab character in the test string without losing focus of the text field?

When I start typing the regex, the cheatsheet gets pushed below the visible part of the page as the "explanation" section expands itself. It would be nice if the cheat sheet remained next to the regex field as I typed in the regex.

I honestly don't know.

I just searched for a few different mutations of the phrase "background transparency" in their bugzilla and didn't see any. I was expecting this to be a case of an end user not taking the time to see if their issue was already in the bug tracker, but if it's in there, it's hard to find. If there were really multiple bug reports (which I'm not saying there weren't) then bugzilla has a pretty bad search feature.

It's a pretty big problem when bug trackers are exposed to a group of non-technical users. Stackoverflow has a nice feature where it does a real time search of existing questions as you type yours in to prevent duplicates. I wish that was a considered a must-have feature of bug trackers.

Shameless plug for a genetic algorithm based trading strategy app I made a while ago. It generates a bunch of strategies with good scores and then the UI let's you pick the one's you actually want to use. The second video is the demo: http://designbyrobots.com/2011/09/06/automated-design-of-tra...

Fascinating to see Automated Design (optimization software applied to product development) in the wild. Some of the more interesting applications deal with situations where the definition of 'optimal' is based on a changing data set, like trending topics on Twitter.

Obviously, this went horribly wrong in this case, which brings us to the lesson:

Optimization technology can generate lots of designs that score well against some set of metrics. You still need someone with decent 'taste' to select which solutions you'll put into production. It doesn't matter if you're optimizing stock trading strategies, jet engines, or t-shirts.

Creating a decent UI for selecting good designs from a giant batch of autogenerated designs is a hard and interesting problem, and almost always needs to be custom built for the domain. Picking the best jet engine needs a totally different UI than picking the best stock trading strategy.

The implication being that the police aren't doing a very good job of contacting people about warrants. The first they're hearing about it is by seeing themselves on Pinterest. Imagine the number of increased arrests if the police started using email.