3 ms·
I really like GigaBlast. I wrote a "meta" search utility for myself that can query multiple search engines from the command line.^1 It mixes the results into
by 1vuio0pswjnm7 5y ago
I really like GigaBlast.
I wrote a "meta" search utility for myself that can query multiple search engines from the command line.^1 It mixes the results into a simplified SERP ("metaSERP"), optimised for a text-only browser, with indicators to show which search engine each result came from. The key feature is that it allows for what I might call "continuation searches". Each metaSERPs contains timestamps in its page source to indicate when searches were executed, as well as preformatted HTTP paths. The next search can thus pick up where the previous one left off. Thus I can, if desired, build a maximum-sized metaSERP for each query.
The reason I wrote this is because search engines (not GigaBlast) funded by ads are increasingly trying to keep users on page one, where the "top ads" are, and they want to keep the number of results small. That's one change from 2005 and earlier. With AltaVista I used to dig deep into SERPs and there was a feeling of comprehensiveness; leave no stone unturned. Google has gradually ruined the ability to perform this type of searching with their now secretive and obviously biased behind-the-scenes ranking procedures.
Why is there no way to re-order results according to objective criteria, e.g., alphabetical; the user must accept the search engines' ordering, giving them the ability to "hide" results on pages the user will never view or simply not return them. That design is more favorable to advertising and less favorable to intellectual curiosity.
Each metaSERP, OTOH, is a file and is saved in a search directory for future reference; I will often go back to previous queries. I can later add more results to a metaSERP if desired. I actually like that GigaBlast's results are different than other search engines. The variety of results I get from different sources arguably improves the quality of the metaSERP. And, of course, metSERPs can be sorted according to objective criteria.
This is, AFAIK, a different way of searching. The "meta-search engines" of yesteryear did not do "continuations", probably because it was not necessary. Nor was there en expectation that user would want to save meta-searches to local files. Users were not trying to minimise their usage of a website, they were not trying to "un-google".
Today's world of web search is different, IMO. There seems to be a belief that the operator of a search engine can guess what a user is searching for, that a user who sends a query is only searching for one specific thing, and that the website has an ad to match with that query. At least, those are the only searches that really matter for advertising purposes. Serendipitous discovery while perusing results is not contemplated in the design. By serendipitous discovery I do not mean sending a random query, e.g., adding an "I'm feeling lucky" button, which to me always seemed like a bad joke.
The only downside so far is I ocassionally have to prune "one-off" searches that I do not want to save from the search directory. I am going to add an indicator at search time that a search is to be considered "ephemeral" and not meant to be saved. Periodically these ephemeral searches can then be pruned from the search directory automatically.
1. Of course this is not limited to web search engines. I also include various individual site search engines, e.g., Github.
- betwixthewires 5y agoWow, do you happen to have published your utility so that other people can play with it?
- 1vuio0pswjnm7 5y agoThe problem is that (1) I am a minimalist and dislike lots of "features" and (2) I prefer extremely simple HTML that targets the links browser. Most users are probably using graphical, Javascript- and CSS-enabled browsers so while this may work great for me, it may be of little interest to others who have higher aesthetic expectations. Another problem is I prefer to write tiny shell scripts and small programs in C that can be used in such scripts. To be interesting to a wider audience, I would likely have to be re-write this in some popular language I do not care for. If I see people on HN complain about how few results they get from search engines, then that could provide some motivation to publish. I am just not sure this is a problem for others besides me. Many results I get from search engines are garbage. By creating a metaSERP with a much higher number of results overall, from a variety of sources, I believe I get a higher number of quality ones.
- betwixthewires 5y agoWell something like that would be interesting to a particular demographic. I prefer minimal aesthetic cruft as well, and like terminal stuff like links. If you ever do decide to publish, be sure to post it here!