Hacker News new | comments | show | ask | jobs | submit login

I briefly liked Quora (partly because its the first site I've seen that serves up minified HTML). Now I find it boring, insular, and full of group think. Much like this article. And the valley. The concept that some non-valley site, that doesn't have all the cool kids behind it, might actually be successful is just not allowed within the techcrunch world-view.

You're a little harsher than I would be but you make a good point. The problem with online reputation on this type of system is most people won't verify the answers before they upvote them. At the same time an insular community creates a competition for status. So you get people who try to find and answer questions as quickly as possible regardless of whether the answer is right or not.

As someone who goes to Stack Overflow regularly I see this situation all the time. It goes like this...

1. Person asks a question (e.g. "I need to build a service that will serve school districts in Los Angeles County. How do I go about doing that?")

2. Someone comes along and answers the question with something that seems right (e.g. "Here are a bunch of links on building REST services...")

3. That answer gets upvoted and the answerer gets his reputation boosted

4. Later someone comes along with the right answer (e.g. LA Unified requires all services to be WSDL discoverable so you need to look at SOAP"). But it's too late because the question's already scrolled into the abyss.

So the guy who gave the wrong answer gets a better reputation and the guy who got the right one gets nothing (and the person who asked the question likely took the upvoted answer and didn't check back so they waist untold hours on the wrong thing)

It's usually not that hard to tell. The answerer from #2 may have a bunch of upvotes and a check mark, but it usually also has a few comments saying "you're wrong".

And who doesn't read the top few answers, if not all of the answers, anyways? Even if you expect the top answer to be the right answer, subsequent answers often help you understand the problem space a lot better.

I think your description of the problem is about right.

But: As a SO-consumer this particular problem doesn't phase me.

The better answers usually have more votes and sometimes the green checkmark is even on the best one - but when it's not, who cares?

To me as a SO-consumer a bigger problem recently has been the duplication. For any given question there may be dozens of threads which makes it harder to track down the good one(s). If SO wants to improve they should imho work on their deduplication efforts.

This might even be automatable to a point; let users vote on questions to be merged, and perform the merge when either the author agrees or a vote-threshold is met.

It's faze. Read more books.

Your pedantry is not welcome here (as supported by your comment score). Please follow this rule in the future for posting on HN: always consider other user's comments in the best possible way they could have meant them.

Wow, urban dictionary. Consider me schooled.

I try and do my part by voting up the answers that are truly useful when I come across a question by searching, but I agree that it does tend to be a first-come first-serve type of system for the most part.

They've taken some steps to alleviate that (badges for getting an older answer upvoted to +5), but it's still an issue.

Having the right answer does help later onlookers into the question.

Moreover, the person asking the question has the most interest, but the least capacity to be patient. And, likewise the initial status-seeking answerers can get edgy for a quick tick.

I'm curious why you're impressed by minified HTML. Usually, given the CPU cost to minify a page and the fact that savings are in the tens of bytes after gzip, it is nearly certainly an overall performance and user experience loss.

I think it makes the browser's job faster and easier. Minifying for me has improved the speed of my sites.

The savings is fairly minimal after gzip, and it comes at a cost -- removing all readability of HTML view source.

Out of interest ... measured how?

Subjective measure. And I am not uncertain if an improvement has occurred or not.

edit: however, I do not minify on-the-fly, I keep the source HTML, CSS and JS files space and line-break compressed (and variable-name compressed.) The speed increase may be due to less going down the wire, or it may just be faster to process by the browser due to some other mechanism.

Also, having css and JS code within the HTML itself has sped things up too - no external files.

Have a look at the source of http://twitya.com/index.php - I also Gzip that, and this speeds it up as well. I keep a few line-breaks in for readability.

Takeaway: what matters most is having as little data go down the pipe as possible in a compressed fashion, and not worrying about server load - that could be offset by the improvements on the browser.

Also, Gzipping http://hackerbra.in has led to a marked improvement, with its large slabs of text.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact