DKB #1 β π€ Welcome to the newsletter & Interview with Brave Search
Hello there!
This is my first ever newsletter, so it might be a little rough, sorry in advance.
I started my new blog dkb.io in January, and by pure chance, my second post entitled "Google Search Is Dying" went straight to the moon π.
I didn't even have an email subscription box at first, but then I thought I should probably do that...since that's what people do, or something π€·.
It felt weird to have an email subscription box, since I didn't even know what the blog was going to be about yet...but I thought I'd figure it out along the way.
A few months have passed, and I think I've figured out what this blog is going to be about, kind of.
Search is a fun topic, but it's only one of the things I'm interested in. For now, the overarching theme of this blog will be "technology and society".
That sounds very generic, but I'm interested in so many subjects (search engines, history, blockchains, economics, education, and too many more to list here) that "technology and society" is the most specific category I can commit to (and even that might be too small a box).
If you're only here for search, then you'll definitely get more of that in the short-term. The next 4 articles will be interviews with the hottest new search engines on the block. After that, there will be more non-search topics.
You can expect new content once a month, maybe, idk. I have a full time job. I'm busy. But I try.
Interview With Brave Search
Today I've published a full length interview with the Chief of Search at Brave. It's interesting to hear the philosophy behind these new search engines that will possibly replace Google.
Here's a small excerpt:
Regarding the problem of deteriorating quality due to spam content, we try to keep it at bay by making sure we index only the relevant subset of the Web. Our index is much smaller than the Web, as a matter of fact it is much smaller than what we could crawl. We index less than 10% of the URLs we are aware of.
We make a conscious decision to index pages only where we have some evidence of usefulness. A small index helps to reduce the garbage-in/garbage-out problem and itβs also more economical to run (still massively expensive).
One way to see it is that we are sacrificing the number of candidate pages to be ranked in return for less noisy results. This is a sensible trade-off, because content also follows the typical Pareto distribution, with 80% of the signal on 20% of the pages.
In any case, the problem of spam, SEO, etc. β letβs call it adversarial content β is indeed a big issue. We might be suffering less than Google by design, but we are not free of it.
Search today is more difficult than it was 6 years ago; infrastructure-wise, we are all better off, but the number of pages keeps increasing and the majority of them are of questionable quality.
Talk To Me
Thoughts, comments, existential questions? Send me an email. I reply to all of them.