Do you mean your database driver doesn’t supply you with field names, that you rely on order to map up data?

I guess I often rely on column order in Go, given Rows.Scan() [0]... so `select ` would be scary since someone may ALTER the table.

Maybe that's the antipattern! But it's also a feature so I feel like you're better off avoiding the . Also, if someone adds additional big columns, you may incur additional, unnecessary cost

[0]: https://golang.org/pkg/database/sql/#Rows.Scan

Your driver can probably introspect, yes. You can also probably introspect yourself if you want to. But all this to be less explicit about what you're querying seems like a waste, doesn't it?

Probably something like:

  to_select = fields - [:created_at, :updated_at]

Though there’s always a slight uncomfort in not being 100% sure the array elements aren’t strings, so might throw in a cast here just to be sure.

Calling it a DSL is bit weird to me, it's a single instance method. You'd use the same kind of thing to set the appropriate caching headers for Varnish.

  subject.should_receive(:popular).and_return(subject)

Now you're testing that your piece of code is calling the "popular" method, but you have no idea whether or not that method even exists in your implementation. Too much mocking is worse than "no mocking", because "no mocking" is slow but it doesn't tell you lies.

Beware that proxies sometimes strip the accept header. I know of at least one European ISP that does this for its 3gs users.

While you basically built this country? It happens that Swedes who are actually paid to think(or paid at all) don't often share your views. This might be because they are not expecting life to be a free ride just because they were born here.

1) It does not follow from "I am not responsible for Romania's poverty" that I claimed "I built Sweden". Läs om, läs rätt. 2) I am not unemployed, nor paid to work with my hands. 3) Even unemployed people and manual laborers can have valid views on things, you elitist. 4) I never once mentioned wanting a free ride!

Here's a helpful two-part Berkeley lecture on the topic: http://academicearth.org/lectures/streams-i

I didn't know that series of lectures, so thanks a hundred times. (Just spend the evening watching the first two hours and bought Harvey's book as well.)

What if the other guys do "Content + Links" as well(seeing how this has been the general consensus SEO silver bullet for 5+ years now), we all win?

Link building is to on-page optimization what marketing is to product development. You can conquer worlds with crappy products and brilliant marketing, but the reverse is also true. Doing both with passion = highest chance to win.

Depending on the type of site and how many links you already have, fortunes can be made by clustering your data/pages better, naming your stuff differently, changing your internal linking etc.. .

"And yes, I think _that_ part of the problem is ripe for automation, and I suspect probably _has_ been automated by the very successful people" You have gotten it backwards. A lot of link building (read: blog spam, paid bloggers, paid articles, link brokers etc..) has become automated, far beyond the level of automation of on-page optimization.