cmbaus

I see you are using SVG to render the charts. What are your thoughts on SVG vs Canvas for these types of apps?

This makes a lot of sense to me. SQLite has the reputation of high reliability and thorough testing, maybe there is a lesson there.

In my opinion the value in testing is verifying what an interface should do, not how it does it.

I agree. Wrapping your library for testing is really pushing the boundaries of sensibility.

Low density code that doesn't provide application logic is one of my pet peeves.

There is a philosophy that started in the 90s (and Microsoft was proponent of it [1]) that adding more layers to an application would make it more malleable, but in a typical CRUD web app, layers only bloat the code and make it slower.

I'd suspect adding a wrapper to a date class just for the sake of testing is more likely to add bugs than remove them.

[1] http://www.asp.net/web-forms/tutorials/data-access/introduct...

It's better. It uses geocities!

That thing is running on Yahoo! Is it an old Viaweb site?

I think there is a trend in UK web design: http://www.lingscars.com/

I just did an 'Inspect Element' on HN. That's some state of the art 1999 web design there.

While the point of the article isn't the domain problem, I'll add that the hardest management job I've had was trying to retroactively fix a data aggregation process that had been built manually over many years.

The work was extraordinarily tedious, error prone, yet business critical. We had to keep shimming in new parts of the system, while keeping the old system afloat. While the system was always on the verge of imploding, from the outside, it basically appeared to work, which kept the project chronically understaffed.

I have a lot respect for anyone who aggregates time critical production data from different sources. The nature of the work makes it extremely stressful, yet surprisingly easy for outsiders to underestimate.

Having worked as a vendor to the financial space for many years, I think it is easy to underestimate how hard it is get off the ground in finance (either banking or money management). There are many obscure regulations, and often the numbers just don't add up. In the end, this probably was a fair deal for Simple.

I think WealthFront is another good example of the difficulty. I like the direction they are going, but they had to pivot many times to get to were they are now, and they are still a fraction of the size of established players in the market. It still isn't clear if their business model is going to work.

With that said, the inefficiency in the money management business is pretty ridiculous. I think companies built from the ground up on new technology will eventually replace existing players in the market, but the margins will probably be a lot less as well.

There was one slide that caught my attention here:

Computer Science: Fixed programs, transient data

Data Science: Fixed data, transient programs

yes.

Understood. It certainly is better to coalesce data with writev() than to use TCP_CORK.

BTW, from my link: "If you need to write multiple buffers that are currently in memory you should prefer the gather function writev() before considering TCP_CORK with multiple calls to write()."

Interesting on the corking stuff. I looked at this a long time ago (http://baus.net/on-tcp_cork/) and found it could make a significant difference for event driven applications.

Personally I think winging it is very difficult to do well. If you know the content very well you can get away with it, but if it is new content, I think it is important to prepare. The best speakers I know spend a lot of time on prep.

Here are couple ideas I've written on the topic: http://baus.net/i-don%27t-like-public-speaking/

I did quite a bit of public speaking in the past couple years and it gets easier over time. I think the best advice is prepare, prepare, prepare.

I remember the article quite well, as it was passed around our office when it was first published. The author gave multiple presentations on the topic around that time.

There is a lot of genius in the article and references, but I will say from experience it is very easy to implement these approaches incorrectly, so unless an application requires the performance benefits provided, it is probably still a better bet to use standard locking mechanisms.

There are some fair points here. The product is more of a futon than a true mattress. It would be interesting if they offered a higher end latex model.

At one point I almost went crazy researching mattresses. I like to know what I am buying, but the industry is non-transparent. The same product can have different names depending on where it is sold.

BTW, another product like this is range hoods.

I'm impressed that you tried to break into this business. I thought the existing companies would have it tied up because of their distribution channels. I did look at latex mattresses at the time, but they were an expensive niche. About the only viable alternative to the traditional mattress industry were the foam mattresses being produced overseas.

Quote from the site regarding his house:

"I wanted to avoid using buiulding (sic) material that tries to look like something it isn't, but with the siding I had no choice: ALL siding materials I can buy try to look like wood! So, unfortunately, my house is wrapped in faked wood. It doesn't look too bad, but I would have preferred something different."

This drives me nuts as a home owner. I don't mind using synthetic materials in my house, but why are are they all skeuomorphic?

Plus I love the way the house is framed with bricks and mortar. I wish more houses were still built like this in the US, but I suspect that labor would make it cost prohibitive.

I agree with the author that having a finite number is not beneficial for a currency -- it is inherently deflationary. I can understand a currency which is increasingly difficult to increase the supply, but limiting production to a fixed amount will certainly lead to hoarding. I've never understood this limitation in bitcoin.

Speculators may get rich in BTC, but high prices do not validate bitcoin as a currency. High volatility is not a desirable trait in a trade medium.

We've done quite a bit of development work in Costa Rica in the past couple years. I think it is an up and coming region.

We have occasional hiccups in San Jose, but mostly it works pretty smoothly. I've worked in Playa Hermosa and had no problems.

BTW, if anyone is considering developers in Costa Rica, send me an email.

I wish I hadn't seen those somnoir glasses. Now I want!

That's true, but the valuations on deals were much lower after the dot com crash and 9/11. Flickr I think went for less than $100million at about the same time, and it was the Instagram of its day.

Outsourcing by eating out has become much more acceptable in the past few years, but I can remember growing up that eating out was considered a luxury. But restaurants can achieve economies a scale that are difficult to get at home, especially if you don't have a large family. In many cases, eating at a restaurant or having food delivered isn't that bad of a deal.

That's crazy that you didn't get the job. Considering AirBNB's growth rate, it seems they could have taken a risk that someone like yourself would grow into the position.

At one point I hired a personal assistant through oDesk. I took this idea from Timothy Ferriss's 4-Hour Workweek, but I hired someone in the US. I wanted to get a landscaping project done (that was required by the environmental regulators in Tahoe), but I didn't have time to track down the contractors, do background checks, make sure they showed up, and finally get the regulators to sign off on the project.

The contractors thought it was pretty strange to have to deal with someone in a different state, but it worked out pretty well. Our culture (at least here in the US) doesn't appreciate outsourcing day to day tasks (and I felt embarrassed telling people I had an assistant). Maybe this comes from our Protestant roots.

It reminds me of an episode of Garrison Keillor's Prairie Home Companion. Some of his characters from Minnesota visit California and find out one of their relatives has a gardener, and they ask if the relative is having any health issues, because they couldn't imagine having someone do their gardening for them, unless they were severely ill.

Update: The poster above mentions getting what you pay for if you pay $2 per hour in the Philippines for instance. I agree with this. I payed significantly more, but I feel as though I got my money's worth. It isn't that expensive to hire in the US if you look to lower cost areas like the Mid-west.

For this project, I felt it was worth it to have someone who understood the problems of hiring and managing contractors in the US. The person I hired was a homeowner, understood what I wanted to accomplish, and did an awesome job. It turned out to be a fraction of the cost of the overall project, and was money well spent.

I think the marketing is creative and well done. Although why would you sell those Alden longwings? Those are awesome shoes! :)

I've outsourced some development on oDesk, and it worked out ok. I even hired one of the developers full-time for a different project.

What I did was start with a small project, and verify the developer's work. This is much easier to do if you are technical.

If you are non-technical, I think these sites are worthless. There is no way to know what you are getting.

Edit: I should say they are worthless for technical work. I've also hired a non-technical assistant who was in the US and that worked out great, and I'm sure other non-technical people could do the same.

Interesting idea.