| 11:35 pm on Sep 24, 2013 (gmt 0)|
Doing search in an entirely different way (probably needing more human input before during or after a search; rating sites, categorising searches etc) suffers from a few big problems. The first is the scale of the problem, it's so easy to underestimate how big and hard things get (and how fast that happens).
There's also what I call the popularity paradox; if something is searched for often enough it's relatively easy to justify investing the time/resources to get it right - but most people who want an alternative search engine want it for more complex/uncommon searches that are not going to be easy to deal with (if they need human input prior to any such search happening) and offering poor results for those will put people off pretty quickly (the people you need as cheerleaders). Handling popular searches well is a requirement, it's not going to get people excited. You need to be good, and consistently so, at things that really matter to people - and by definition it's often niche searches in popular subjects or pretty uncommon searches in general that are really going to impress people (if you wanted to test a new search engine I'm sure you would go for a few test searches in popular areas, your business area, do some local searches and do searches in niches that you have a lot of knowledge in - plus a few others). Part of the problem is the fact that many searches are "new" in the words that are used (and we're a long way from being able to automatically spot what people "mean" - that's part of what's being complained about).
I honestly believe that it's going to take something special to change the way people use search. My money (and time) is on beating current search engines in areas that are non-commercial for them (or should be non-commercial) and then moving into areas which are commercial but tackling those with less concern for profit and more concern for precision.
Some searches should not be influenced by the bottom line at all, that's not going to go down well with shareholders. The next big thing NEEDS to be different, it needs to follow something like the Social Enterprise model (as in the UK definition of that where it means a company that re-invest profits in the project, not as in social media - although it would need to build a community).
I predict that we will see this tackled vertical by vertical with a common platform being used so that uniting results later for general search is possible. Who would build such a platform? Where would it start? How would it be funded? Maybe we will find out soon...
| 1:37 am on Sep 25, 2013 (gmt 0)|
To start with, Cut off all outside sources for premium content/widgets that would try to "index it", including G,M,Y, hell make'em pay for access at a premium, sort of a "scrape page per" page for [insert your currency sign here]0.99, then charge an obscure amount higher that what they claim it to be in current PPC Rates on their own serf, if less that +1000 a day, then sorry have to pay a fee.
It would be so easy to flip big G on its belly if a vertical, one at a time after vertical, simply puts the wall out, pay or scrape, feed the beast, seek beast at the beast the way they did it to us. Ye I am talking to you.
If none paid search(you know yo niche, u been at it for a while ah?) Include only the content that has been voted by the vertical members(exclude current social network gaming theory to start with), open source the thing. Exclude fluff by big box unless quoted as reference from with in.
Oops... I White-Listed again.....
| 2:30 am on Sep 25, 2013 (gmt 0)|
|I Want a New Search Engine |
Can I colour the pages of the design when we reach a collective agreement on how we're going to build this?
| 6:08 am on Sep 25, 2013 (gmt 0)|
The problem with asking to get Google from 2010 back, is that they cannot cope with the spam. You can get Google with a simpler interface and just the SERPS (Startpage), but you still have spammed SERPS and the effects of Google's attempts to counter it.
| 10:09 am on Sep 25, 2013 (gmt 0)|
Blekko is just as good as Google, give it a try:
[edited by: lawman at 11:12 am (utc) on Sep 25, 2013]
| 10:16 am on Sep 25, 2013 (gmt 0)|
Another good search engine:
| 3:03 pm on Sep 25, 2013 (gmt 0)|
|Maybe the current search engines don't see any profit in that, but there is such a big niche that would try an alternative if it was properly marketed as a whole new way of searching the net. |
DuckDuckGo adds value to Bing results by taking Bing's data, running it through proprietary software, and serving it up on DDG pages. Maybe Google could try something similar with its own results data. What if Google offered an advanced search application (I like the name "Google for Grown-ups") that would let users or third parties devise their own SERP layouts and filters? The application could even have "search sets" (shopping, information, etc.) like the tabbed pages that londrum proposed.
Conceivably, this could make space on Google's SERPs more valuable to advertisers, because advertisers could buy "shoppers" (for direct-response ads), "information seekers" (for branding or awareness-building display ads), and so forth.
| 6:46 pm on Sep 26, 2013 (gmt 0)|
EditorialGuy, great point. I've also wondered if it might be possible to let browsers refine search results according to the user's set preferences. Browser setup could guide you through this, and it would be a new and exciting "feature" for browsers. Not sure that's feasible, just throwing ideas out there to keep the ball rolling.
| 10:36 pm on Sep 26, 2013 (gmt 0)|
Have you tried Clusty dot com? They do offer options to search within or block results from videos, web, news, wikipedia, blogs, jobs,shopping or .gov sites. You can make all kinds of preferences permanent or this search only, how many results, how many clusters - search only for pdf, xls or other extensions. Probably why it is not the most popular in the world, but they do find results that others ignore. They offer tabs and you can custom name them. I don't use it all the time just because I prefer Ixquick, but it does a lot of things that others don't.
| 8:07 pm on Sep 30, 2013 (gmt 0)|
Oh, I'd be somewhat content if only the words that I wrote in the search box were actually present on the pages listed in the SERPS.
I think todays search engines -- or at least one specific such -- have been made so technically advanced that they no longer get the basics right.
| 9:05 am on Oct 1, 2013 (gmt 0)|
They are not technically advanced, they are commercially advanced. The results right now are very poor. Search is being defined by advertising as opposed to results.
If a decent research search engine was to appear, one that people could be taught to use with filters and operators, it could be adopted by schools and educational establishments. This would lead to many people continuing to use it and less reliance on Google. Google's search engine is designed largely for the brain dead.
There, I said it! :)
| 1:31 pm on Oct 1, 2013 (gmt 0)|
Q: When is the last time you looked under your car hood?
So why should I see a search "engine". The best search engine should never been seen. It should be an extension of an OS.
> he wants google from 2010
No, not really. I have not seen a utility I want yet. What I actually want is a toolbar utility to do all this for me.
> how little is your time worth
My time is worth alot - hence wanting a search engine that does its job and "searches for me" works.
> an engine that asks you to give more input and
> rewards you with more refined results
Ding ding ding - we have a winner.
>Can I colour the pages of the design
No, because the best search engine will never have a 'page'. It will only be an extension of the OS.
>that they cannot cope with the spam
I rarely rarely run into what I would call 'spam' except by Amazon, HomeDepot, eBay, Yelp, or Wikipedia.
> let browsers refine search results according
exactly. The next good search engine is just an API service run via local browser. It's time to take the WWW out of the search 'business'.
| 5:20 pm on Oct 1, 2013 (gmt 0)|
Marry search to the O/S, and the machines are your in-laws... no thanks. Patched pieces have weaknesses for sure, but monoliths too often become monopolies.
| 8:08 pm on Oct 1, 2013 (gmt 0)|
But Brett's not asking for search engines to go away completely. He's asking for an alternative technology. You wouldn't have to use it if you prefer the engines.
|brotherhood of LAN|
| 1:49 am on Oct 2, 2013 (gmt 0)|
There's surely room for a paid for service. Flat rate or per search, say 0.25 cents a search.
cuil had the kind of scale that'd be required, the publicity, but lacked relevancy.
$50 million and relevant results should get you what you're after!
| 9:06 am on Oct 2, 2013 (gmt 0)|
There is room for an ethical, not for profit search engine designed to find the results that people are seeking with no commercial pressures to influence these results or how they are presented.
| 9:58 am on Oct 2, 2013 (gmt 0)|
@Brett_Tabke, you do not see spam because of the reliance on known, safe, big sites (the sort you want eliminated by a "big sites suck mode") in order to beat spam.
I do not see what the benefit of web search integration with the OS would be - and we already have search integrated with the browser. I never go to search engine home pages.
@BeeDeeDubbleU, who will pay for a not for profit search engine? They are expensive things to run...
| 10:48 am on Oct 2, 2013 (gmt 0)|
They would not now be that expensive given that much of the technology is widely available. One obvious way of raising funding is to ask commercial companies to pay for listing.
|@BeeDeeDubbleU, who will pay for a not for profit search engine? They are expensive things to run... |
It's the only way we will get search that is truly designed for and by the user. I started promoting the idea of a not for profit search engine about eight years ago in here and I was ridiculed by some folks for suggesting this. I suspect that some of them may now see where I was coming from.
Rather than repeat myself (and if you are interested) have a look at the discussion that starts at message number 251648 at the bottom of this page from January 2006.
| 7:03 pm on Oct 2, 2013 (gmt 0)|
One of the big problems I have with Google is the automatic stemming and synonym matches. I hate it.
| 3:29 pm on Oct 8, 2013 (gmt 0)|
> I never go to search engine home pages.
So why would you want to go to a SERP? Why not just open the results in 10 tabs and walk through them. The snippets are 99.9% useless. It takes longer to read a title/snippet than to click and get the site to open.
How I search:
- enter search in address bar
- results pop up
- try to avoid 'amazon', 'ebay', 'etc'.
- try to avoid duplicate domains.
- shift click each result into background tab
See, I spend more time with the snippets trying to avoid stuff, than click because a snippet is interesting (they never are).
| 4:03 pm on Oct 8, 2013 (gmt 0)|
after all this panda/penguin .... (insert whatever word here)
i made a system to analyze: code-to-content | images (inc size and content of img (if text present then relevance to page)) | dup content | IBLs | OBLs | METAS | ETC ETC
what i realised is that it is the basics (and i mean basics) of an algo that could be 12% (maybe more) of a search engine algo
what i would *like* is an open source SE, especially for the more academic searches
if i want to shop i know which online stores i trust, albeit sometimes i may search for a certain vertical
i like the "no serps" idea but it would mean that sites (at the backend before being presented to the user) would have to be checked for the usual nasties (adware | spyware | etc)
i remember when the big G first surfaced and how refreshing it was. i do long for those days as i am sure every webmaster (except holders of G stock) does.
to summarize, yes brett i agree completely (to the majority of your *wishes* if only we could make it happen!
| 4:11 pm on Oct 8, 2013 (gmt 0)|
But then perhaps you are not typical.
| 4:16 pm on Oct 8, 2013 (gmt 0)|
"But then perhaps you are not typical. "
is that to me BeeDeeDubbleU?
| 4:36 pm on Oct 8, 2013 (gmt 0)|
No, to Brett. :)
| 4:37 pm on Oct 8, 2013 (gmt 0)|
ok! sorry! too much caffeine!
| 4:59 pm on Oct 9, 2013 (gmt 0)|
I would prefer better snippets to opening up multiple tabs. I tend to have too many tabs open at all times already.
| 12:42 pm on Dec 6, 2013 (gmt 0)|
I was ready to throw the bleep machine through the window, I was looking for something specific to my activity and got only sponsored link but no real meat!
then I thought to revisit this post, tried the same search with Blekko and then Eureka found what I searched.
I perform probably close to 50 searches daily, guess I will use Blekko.
However I do not know if there is such a thing, but I will patronize a for a fee SE, I might pay from 20 to 50 monthly upon kind of range of services offered (no add served).
| 5:13 pm on Dec 7, 2013 (gmt 0)|
I'm liking duckduckgo these days. Seem to be finding much more relevant stuff.
| 5:19 pm on Dec 7, 2013 (gmt 0)|
I'd agree that the quacks often times know best. I jump to them when ixquick fails me. By the way, I'm still in a grumpy mood for not being able to colour the new pages of the design. I was just listening to Meatloaf, Bat out of Hell, and was thinking that would make a pleasant background image. Bat boy riding a Shadow 1100 in a hurry!
| 5:33 pm on Dec 7, 2013 (gmt 0)|
Hey Brett, how 'bout if I can't colour the pages I can name the slogan...
"We'll Deliver the Truth to You at any Cost"?
| 3:22 pm on Dec 8, 2013 (gmt 0)|
I learned a lot just from the question and posts made. Thanks!
| This 61 message thread spans 3 pages: < < 61 ( 1  3 ) > > |