[Disclaimer: shameless plug, and also previously announced as a Show HN a while back]
I would like to mention my search engine as another privacy-focused alternative: https://solveforall.com/
1) Does not track user activity. Hosted in Canada.
2) Does not leak referrer to visited sites
3) No ads. Will be considering affiliate links, a paid API,
and/or "good" ads -- ads people want that don't compromise privacy
4) Integrated feed reader which also provides search results
5) Activation codes (like DDG bangs, so ?g instead of !g)
5) Plugins written in JS/data to be searched can be added at any time.
6) Deep search -- get results from the search results page of several sites at a time. Try https://solveforall.com/answers.do?q=rx+480&client.kind=web&...
There clearly a lot more work to be done, and I plan on open sourcing this soon, but please try it out and let me know any feedback you have!
As a San Diego resident for the last 12 years, I have to agree it is not a good place for a tech worker. Salaries are quite low compared to other areas, companies aren't very many or particular successful, and housing costs in the good school districts are high (at least $600k for a SFR). Public transit is awful, and car traffic is quite bad as well.
If I didn't have family here, I'd definitely move.
I have found San Diego challenging if you want to work for a big company with a good salary or get a contract at a reasonable rate. There are some startups in downtown SD though. Commuting is pretty tough without a car, unless you live near one of the few rail stations. And the trains are slow. If I were young and unencumbered, I'd definitely be in Seattle.
Really helpful feedback, thanks! It is true that the focus is on niches, like programming, deals, and politics. I'll be working to incorporate your feedback.
About the font, I think my taste in fonts is pretty much opposite of everyone else's. I wanted to do something more "stylish", but I guess I'll switch to something more professional. If you happen to have a suggestion, or just general guidelines, I'd love to hear it.
I assume your "try after google" list is figurative, not a real list on the web?
Ah, I get it now. Thanks for the explanation. This is the problem with the "deep web" not being indexable by crawlers. I've started looking at how to make this easier for users. Basically I'm got some ways to detect which sites should be searched in response to your queries, based on the category of words in your query and the site. But then I need to do a real-time search of that site, that might require JS to run, extract the links, and present them back to the user. It might be slow, but it's easier for the user to wait for the cloud to do it than click the links on multiple sites himself.
Thanks for checking out my site. If you have any questions or have any trouble please let me know at [email protected]. And maybe we can explore brainstorming/collaborating since you've clearly put a lot of thought into search.
To be fair, when I search I get a huge Knowledge Graph result on the right with a picture of the animal.
Generally, I think that the number of people who want the Urban Dictionary result is higher than the number of people looking for the animal, so it's hard to fault Google for this.
What would be ideal is to have something like DuckDuckGo's disambiguation bar that included the Urban Dictionary definition of goat:
In terms of results from crawling, I would consider Bing a "serious" competitor, although still a distant second. They are well-funded, used in ~30% of searches in the US, and don't look like they're going away anytime soon. It would be sad if they did because Google would then be a total monopoly.
Though DuckDuckGo does do its own indexing, AFAIK it is limited and they mostly rely on other search engines (Bing and Yandex most probably). So it's more a meta-search engine not quite in the category of Google.
Today, I just posted a Show HN for a search engine and feed reader I've been working on. It also has "bangs" except they are called activation codes and start with a ?, like
"?jq appendto". It's easy to add your own "?" handler, as it doesn't require any approval to do so. I'm just getting started so I would love to get any feedback.
Can you explain the image and what it's trying to show? That the number of text links should be more than the number of image links to well known sites? I'm asking because I do something similar on my site, solveforall.com, but to a much lesser degree of course.
Thanks for the feedback and the link! To answer your questions:
1) "I wonder what does it mean by "customizable""
You can add different plugins (that produce answers) and control when to activate them. You can also create mini-search engines that search specific domains. There are also a few settings about security that you can tweak.
2) "how is this different from DuckDuckGo with its duck.co"
This search engine also includes a feed reader and get results from your feeds when you search. Also, the plugin system allows you to add data and plugins (written in 100% Javascript) instantly without anyone's approval.
3) "Why I should register for an account?"
Since this includes a feed reader, you need to create an account to save your chosen feeds. Also by creating an account you can create your own mini-search engines and add plugins other than the default.
Since you've asked these questions, it's clear to me that I haven't done a good job explaining these things on my site. I'll be working on improving this. Thanks a lot for the feedback!
Hi, I'm the developer of Solve for All. I'm hoping to get feedback on this project, which I promise to consider carefully. I'm also hoping to get in contact with anyone interested in working together, sponsoring the site (which is ad-free), or any other reason.
A few caveats about the site:
1) I've made the settings a little less private for the sake of this Show HN. I don't share your search history with anyone, but with the current defaults, you'll be downloading images directly from the sites in your search results and feeds, possibly over HTTP. When I finally have a 1.0, I plan on securing all downloads for subscribers.
2) Sorry about the captcha when you create an account. Almost always the captcha is impossible when you first load it, so you have to hit the blue refresh button. I couldn't find a good solution that didn't require loading a 3rd party script (Google noCaptcha seems to work beautifully, but I'm wary of including their script). For now I'm using a locally downloaded copy of the reCaptcha script.
3) I'm not wedded to the name, logo, or the theme. Feel free to criticize any aspect of the site, I can take it!
4) I'm specifically looking for feedback on how to make this open source, but still making this a sustainable business. I am thinking of AGPL, or possibly something that would prohibit commercial usage without a license.
Title is certainly weird -- his rant did not cause the user space to misbehave, the bad commit did.
Seems like he is having a very bad day. Still, I'm glad I'm not skilled enough to even attempt to work on the Linux kernel. I feel sad that others might be scared away, but maybe that's keeping the quality up.
I had a hard time using this product. I was trying to use another developer's APIs but it didn't seem possible. The one positive thing about the announcement, from the FAQ:
Will Palantir (or anyone else) have access to my personal data?
No. All of your user data including names, email addresses, passwords and APIs will not be shared with anyone and will be securely purged from our servers on March 31, 2016.