leeleelee

I like to use python, flask, gunicorn, and nginx for back end apis.

Now that I've been doing so much front end javascript programming, I'm looking into Node just because I'm way less efficient at writing python and more efficient at writing javascript nowadays.

At least that's my personal decision criteria.

It might look kind of cool at first, but it puts additional barriers in between the user and actually doing something useful like entering data in a form, booking a flight, etc.

For example, if I haven't been to the site before, I might have to explore the interface a bit before I get to the part of the interface I want. Clicking on those tiles makes things shift and bounce around, the content portion of the screen constantly moves left to right.

Way too much difficulty for doing something so simple.

Designs should be visually appeasing, but not at the expense of ease of use or ease of understanding.

Ah, you are correct. I missed that.

This is a premature conclusion. There was no measurement of IQ before and after infection, so no such conclusion can be made.

There could be other, unknown, factors that happen to affect BOTH: a person's IQ and their likelihood to be infected, along with severity of infection. For example some genetic, environmental, or other variable that affects your IQ as well as infection-resistance.

It very well might be true, that infection does have the ability to lower IQ. But this study does not prove it.

I wish scientists would be more straightforward about conclusions. The title of the article should read: "Study finds that individuals who have suffered infection have a lower IQ than the general population" or something along those lines. Much different than saying that "infection lowers IQ".

This is beginner-level confusion of causation and correlation.

Glad to see articles like this, I've had the same point of view for awhile now.

At least uber has revenue though. Lots of companies are getting funded with zero revenue (nevermind profit for now) and no clear revenue model. Crazy in my opinion.

Last time I did was at least several years ago. The experience was terrible enough that I ran and never came back. To be honest I don't remember what exactly I had to do, but it wasn't as simple as following an install wizard. Lot's of little hack-ish type stuff.

Has anyone gone through the horror of doing things like installing python or mysql on a windows machine? Maybe it's easier nowadays, but last time I tried installing python I had to follow multiple tutorials, one of which actually worked. Installing mysql was a little more pain-free, but still too much work.

Also the lack of a linux/unix/whatever shell in windows? There's no way to measure the infinite utility in having a shell at your fingertips on a mac. I use the shell everyday, not sure how anybody could get on without one.

A lot of people probably run into problems from poor design, poor programming, or just inability to problem solve. For example (two real examples I've faced):

(1) Company xyz is having problems generating reports that pull from an sql database. They take too long, IT says it's because the data is too large and their hardware is old. Obviously this is big data! Wrong, re-writing the sql queries and indexing some fields solved the problem. Like, beginner-level stuff. Nobody had even looked at the code or structure of the database when initially trying to solve the problem.

or

(2) Company xyz can't load a dataset into memory to perform some computations on it, it's too large. Needs better hardware or big data solution obviously. With millions of rows and many columns, ok. Only 6 of those columns were even used by the processing job, and a fraction of the rows were relevant. There was no need to fit the entire dataset into memory in the first place, problem solved.

I could go on. Lots of companies just lack the skills and knowledge to do things right, and have no idea what's going on under the hood or how things work. So naturally, they are drawn to the marketing buzz surrounding big data, cloud computing, data warehousing, machine learning, etc etc etc.

I've never had a problem implementing realtime search. Can somebody give me an example use case where algolia is a better solution than doing it yourself?

Clearly it's a popular, in-demand service -- I'm genuinely curious as to the types of apps/websites this is best suited for.

Very, very awesome article.

I never used Secret, but I can't imagine how an app like that would ever make revenue.

Maybe I'm just not creative when it comes to figuring out how to monetize these silly social apps.

Can anybody explain what goes through the minds of investors and founders? How do you eventually make revenue with something like Secret?