I heard people comment on all the Twitter & FriendFeed real-timeness, pointing out that Google is fast too. But maybe they are just crawling FF and Twitter to look real-time?
- Ruud van Wijngaarden
Ruud: friendfeed items get ranked higher in Google. I've seen that happen a LOT.
- Robert Scoble
I noticed the same thing on Mahalo Answers. In less then half an hour the question shows up in Googles search results. Google is just crawling really fast.
- arjo
Google was really slow earlier today with the Jay Bennett/Wilco news. Couldn't find anything on Google, it was all over Twitter for hours before it showed in Google.
- Karoli
Maybe the FF structure or API (and from Mahalo and the likes) makes it extremely easy to index for Google, so Google is actually preying on the smart systems of others to be sort-of-real-time?
- Ruud van Wijngaarden
Well, THIS conversation isn't in Google...yet...
- John E. Bredehoft
Actually this conversation is on Google already. If you search for "Wow, is Google going real time already?" it's there.
- Charbax
Since Friendfeed provides a real-time API, why wouldn't Google basically store a searchable copy of all of Friendfeed on their index?
- Charbax
Charbax: not to mention that friendfeed is more SEO friendly. The URLs here have keywords in them. I'd expect that from a team of Google superstars.
- Robert Scoble
Do you really think Google's entire search engine can transform into real-time...I think it would take at least 3 years, if at all. Friendfeed engine is very friendly but substantially smaller, with only minor queries compared to the big ones. Google might improve its integration system but the path to a frienfeed real-time module looks much longer than expected.
- Nir Ben Yona
I think the structure of Google, and all old skool search engines (searchword to your mother!), are not compatible with real time data gathering. Their bots would probably slow down the whole web if they were everywhere all the time. The nothing will be real-time anymore...
- Ruud van Wijngaarden
Rather than thinking "the burden is on Google to spider the web in real-time" think "clever sites with active communities can help Google give the appearance of real-time web search." I don't think people blog in real-time, but they do in the aggregate, and some sites do better than others in helping out the "spiders."
- Richard ¿digame? Walker
Strangely, Robert's search link doesnt have a single entry from Twitter...atleast in first 3 pages i checked.
- Roshan Ramachandran
Roshan, look at first link in the screen shot I captured right above - It shows a twitter reference ... Its a direct link to his twitter page ...
- Brent - Long Live Rock
It makes sense since this item originated over on Twitter.
- Robert Scoble
The other one is mostly showing friendfeed, because that conversation started on friendfeed. Interesting how Google works.
- Robert Scoble
http://blog.broadbandmechanics.com/activit... - this is getting a little weird. Here there is a link to yoiur twitter page which is alright, but then all the friend feed chat is below your Link ... WTF? and it is in real time ... Ok apparently this guy is taking a major chunk of FF Feed and posting it on his blog in real time. I'm not sure, but it sure seems like plagerisim to me ...
- Brent - Long Live Rock
I agree with Ruud, Google indexing social media sites (as well as links) is an effective way to appear real-time, as they are already where most of the action happens anyway. Some sites are always more likely to contain current content while others are a waste of resources to crawl more than once a month.
- Alistair (alpinefolk)
I imagine google could also identify frequently updated sites by making notes about how often each site is updated, each time the bot passes through. Then send the bots to rapidly index sites with a high real-time score. It could be a self-adjusting process that would ensure no resources are wasted rapidly scanning static sites.
- Raj Advani
exactly Raj, it's a constantly self-adjusting algorithm that can distinguish sites that are "static" or "dynamic" or "should be dynamic but aren't at the moment" (i.e. tube blockage) - and also perhaps predict spikes and schedule the site's time slot for the best spidering payoff.
- Richard ¿digame? Walker
Adding the Greasemonkey script "Twitter Search Results on Google" combined w/ an advanced Google search makes good comparison
- Mike Elliott
I think this is funny/cool - I don't normally see my stuff show up for a while but to have it show up as quick as it did is interesting. That was a fun conversation last night - I love Web 2.0
- Robert Freeze
Yeah! It seems like they've tweaked their servers to hunt after these realtime mirco-blogging sites.. Even, Tweets are showing up early like never before. :)
- Mohammad Abdurraafay
It hasn't come up in the discussion yet, so I'll mention ... is it possible that many of you are getting unrepresentative results, if you have recently visited a site, and you then Use Google to search for keywords or phrases. Google may be returning results to YOU, individually, more quickly, by using your recent browser history to return customized Search Results based on sites recently visited.
- Gary Etie
... Also ... I get Google Alerts, via Gmail, 10 minutes after posting to my blog, if the post contains Google Alert search terms that I monitor. Getting a Google Alert, 10 minutes after a self-hosted, WordPress Blog Post brings other factors into the "real time search" discussion, like the affect of "Notifications" that go out to Google, MSN, Yahoo, etc., via WordPress Plug-Ins, as soon...
more...
- Gary Etie
Gary: Google treats everything as equal. It's highly unlikely that they are customising search results well not yet anyway - this will affect their algorithims. I think Google are just weighting the content on FriendFeed / Twitter more important by indexing the /everyone feed every five minutes or so instead of daily etc (which can be set by a webmaster via webmaster tools and a sitemap.xml ;))
- Nicholas James
Really? Google doesn't customize Search Results to the individual user, based on browsing history?
- Gary Etie
Search Wiki is be customized (although the user is editting the results themselves) other then that users are treated the same.
- Nicholas James
Its not just searchwiki that's customized and adding &pws=0 does not come close to removing all customization.
- Ryan Underdown
from fftogo
I posted an entry on my blog ( http://austincitypermits.com/blog... ) with the words "Current Google Gap in Real Time Search" in the title. Time stamp - May 26, 2009 @ 21:40. I'm going to Search Google periodically, to see how quickly it gets indexed, and post the timestamp of the Google Alert I receive.
- Gary Etie
Can you tell me why "real-time" search is good?
- Jason Nunnelley
I believe it's good for everyone in case you happen to be looking for the information. It's good for Google because if they can't do real time then the twitter search and everyone elses search that does will dominate more and more.
- James Stratford
Googles plan of Real Time Search at the moment is linking to the conversation...we'll see this shift eventually.
- Nicholas James
Time stamp - blog post - May 26, 2009 @ 21:40 .... Time stamp - Google Alert - "austin city permits" - Tue, May 26, 2009 at 9:52 PM - That took 12 minutes.
- Gary Etie
A normal Google Search already has the post at #1 position, due to exact wording. It took me 30 minutes to get back and do a search, and it was already up at 22:10 PM CST ... http://www.google.com/search...
- Gary Etie
Robert - I've done a few blog posts on this. I call it 'hyper-indexing' ... and they've been doing it on select blogs for over a year (A lot of blogs ....)
- Charlie Anzman
Charlie - That's precisely what I'm attempting to point out. My blog is not select, by any means, and I get a Google Alert within 10 - 12 minutes after a post, and the post shows up in Google Search results within 30 minutes.
- Gary Etie
equal is not good ,but snap internet in 1s is cool
- qian
Yes I have noticed this too in the last few days
- Nova Spivack
so that explains why I got so many vanity google alerts today.
- Peggy Dolane
after forgetting that I had joined Twine I was reminded by a post here at FF and decided to give it another try out (http://www.twine.com/user...). the one thing right off the bat is that it seems 'slow' but then I think we get spoiled with things like FF
actually I wasn't thinking of the real-time aspect Hutch but rather how well FF loads in the first place
- Steven Hodson
Steven: I agree. FF has always been fast for me. I load it hundreds of times per day. To me this is a major requirement for services I'm going to use so often.
- Robert Scoble
Steven, what do you think about the complaints that twine copies too much of original article reducing pageviews?
- dthree
@David give me a couple of days to opine on that as I'm still getting me feet wet over there :)
- Steven Hodson
Yeah FriendFeed does load well. Twine has a lot of algorithms behind the scenes it has to compute, doesn't it?
- Hutch Carpenter
Twine will get faster -- as fast as FF -- in the next release (T2). New improved architecture will solve that. That is still several months away at the earliest. Meanwhile, it does do a lot of computation, and has grown faster than expected.
- Nova Spivack
And by the way, Twine does not compete with FF -- they are entirely different animals.You can track Twine in FF in fact.
- Nova Spivack
I will probably add it to the mix here as well once I get the hang of how things are done on Twine .. kind of looking forward to the exploration
- Steven Hodson
This should be exciting. Remember, Wolfram discovered cellular automata, too.
- Akiva Moskovitz
So long as the beta isn't called Wolfram Hart, it'll all be good.
- WorldofHiglet
Akiva, I believe it was actually John von Neumann (in '40s...working on self replicating systems he followed suggestion by Stanislaw Ulam to use mathematical abstraction rather than having one robot physically build another robot). For those of us in school in the early 70's our first exposure to CA was via an article in Scientific American (Martin Gardner's Column) on John Conway's "Game of Life." Wolframs' work was considerably later....(I think first published around mid 80's) but oh what work it was!
- David HC Soul
Sorry, Akiva but Wolfram didn't discover cellular automata von Neumann did and the NYT article from several years ago was talking about Wolfram playing with the Game of Life created by John Conway in the 70's.
- Jimminy
I was being sarcastic. His whole pompous 'new science' deal and all that.
- Akiva Moskovitz
Sorry Akiva - right over my head; it must be too late at night for an old guy like me.
- David HC Soul
Looks like it will be a giant, general-purpose expert system.
- Pavlo Zahozhenko
S'all right. I'm barely keeping it together myself.
- Akiva Moskovitz
My goodness - #WolframAlpha news is incredible. I've been so looking forward to hearing about something like this.
- Julian Edward
Robert, I too want to see this... the potential for this is really amazing. Einstein would go nuts for this, huh?! This is a great milestone in mathematics and computing science
- Susan Beebe
Hi folks -- it is indeed very cool -- I was blown away by what I saw. It is really something new and impressive. Only Wolfram would take on something so ambitious, and actually pull it off.
- Nova Spivack
Hmm... what's stopping Google from creating something similar? I guess because of what the article highlighted: hidden in the article is a big red flag for me, " ... there are potential biases in the answers one might come up with, depending on the data sources and paradigms used to compute them."
- Pandu ● IT Optimizer
I have to add a note: Ever tried searching for "who is the mother of britney spears" in Google? Try it.Or, for a more acceptable kind of question, try "what is the capital of bolivia"
- Pandu ● IT Optimizer
Pandu I believe that the idea of Wolfram is to answer other questions i.e. "What was the level rainfall on X day in X". Also Mahesh they're is a conversation going on over on Twine which explains the reasons why it is different in the comments.
- Nicholas James
I've noticed a pattern over quite a few years: Internet services that are pre-announced with a great deal of hype often fail. Services that simply and quietly open up shop, like Google, and which rely on word-of-mouth to communicate an exciting experience, are more likely to succeed. NEVER pre-hype your Internet service. Let the service speak for itself. That being said, from its...
more...
- Sean McBride
If it takes that many words to describe it I'm thinking it's not going to pan-out.
- Kevin Gamble
I dunno, I remember google having a good amount of hype long before I ever used it, I don't think it did "pre-launch" but certainly long before it was well known and at a time when other search engines results were still competitive.
- Richard Lawler
Have any pointers to the Google pre-hype? I don't recall it.
- Sean McBride
This all seems to rest upon some interesting advances in Knowledge Representation, which might be of use even if the service doesn't pan out.
- Seth Greenblatt
Where are the research papers underlying the knowledge representation approach here? Wolfram's team worked entirely in stealth mode?
- Sean McBride
I have a friend working on this and he is raving about it.
- Mark Krynsky
Mark -- where is the headquarters for this project?
- Sean McBride
reminds me that we still have no clear picture of how mind works.. and i don't think we could ever possibly have..
- Hayk H.
Hayk -- we know quite a bit about how the mind works, and will learn much more. Cognitive science is a viable endeavor.
- Sean McBride
Sean, that is true. But compare the amount of money, resources, time and effort put into finding out how the mind works, combine it with all previous endeavors of humanity and contrast it with results we have - IMO not much. Processing info in as organic and natural a manner as human mind is the objective of this project. I am going to be cautiously skeptical in my expectations about it, although I already see the hype that gathers around it - reminiscent of Turing's machine some 70+ years ago.
- Hayk H.
Sean, my friend is based in LA but I don't have details on where the rest of the team / offices are located.
- Mark Krynsky
Where Sergey Brin once said "If it doesn't exist, you cant find it" Wolfram is saying the answer can be constructed from the question. Notions of what 'does' and 'does not' exist are going to be pulled in for questioning.
- David Bausola
Wolfram and his team are smart. I look forward to seeing what impact this has on the community. I wish that he would do more work in "productionalizing" his research in emergent behaviors.
- Will Hawkins
I had problems before. When using the bookmarklet you have to select a twine as a destination, or it won't show up here. I hope that helps.
- Alejandro
Alejandro, no I've added it as a service in my Account page, so it should feed into FF automatically but nothing at all. :-(
- Kol Tregaskes
You know, I did that, and forgot all about it. I shall go and check it out now you've reminded me. So much Social Media, so little time!
- Ian May
Anything you post to any *public* twine appears in FriendFeed -- start your own personal public twine... like a blog ... or post to others you join
- Nova Spivack
This is still not working for me, I have create d public twine and added 2 articles to it but nothing is coming through. :-(
- Kol Tregaskes
You can pull the rss feed from your public twine into friendfeed like a blog, or you can just add your twine account to friendfeed to get all public posts you make to any twines.
- Nova Spivack
Ah OK, but then what is supposed to be pulled in when you add it as a proper service in FriendFeed? It seems pointless to have it as a service you can add in FF if you then have to add any Twine feed as generic blog. :-(
- Kol Tregaskes
I wish Twine could import my delicious items automatically instead of making me do it manually. I like the tagging functionality in Twine and might even use it more IF it automatically picked up my items in Delicious as I add them in Delicious. - http://www.twine.com/user...
Dude -- why even bother using delicious anymore? Switch to Twine. WAY more powerful and a better long-term investment in your data's future. Of course I am biased... But I'm also right. :)
- Nova Spivack
Merely saying it doesn't make it true. How is Twine more powerful, if it can't interact with other web services? How is it a better long-term investment? If I want my data out, can I get it? In what form?
- D0r0th34
It's better because Twine is soon going to be datamining all the links you add and gathering the full text and building you a search engine from them. You can get data out by RSS already. Full API coming in early 2009.
- Nova Spivack
Simpy already has an API, and you can build yourself a custom Google search engine any time you want. Just sayin'.
- Bill Hooker
I signed up for Twine (quick 'n' painless, nice) -- it seems pretty fast and easy to navigate (I wouldn't say intuitive, but learning curve not too steep), but until I can import my bookmarks from elsewhere I won't even be considering a switch. There's a FAQ item (http://www.twine.com/item...) that seems to indicate that bookmark import is functional, but I can't find it...
- Bill Hooker
I agree, until I can get my 3+ years of delicious bookmarks in, there's really no use. Plus, is there a firefox extension that matches or exceeds the Delicious functionality? I love being able to control+B, type in a phrase, and then arrow down to the page. Also, I miss DeliGoo. :(
- Kårín Dalzĭel
Trying to use RSSFWD to send my delicious feed to my "post by email" feature in Twine. Perhaps this will auto add my delicious bookmarks to Twine. Waiting to see if it will work. If not, will try xFruits instead of RSSFWD to see if that will work with Twine's post by email function.
- Stephen Francoeur
I have been using Delicious for years and it does exactly what I want it to do. I like the simplicity of function and design. When I look at the Twine site, I'm not exactly sure what it's trying to do.
- Steve is older than ever
Doesn't look like RSSFWD is working. I thought I'd take my delicious feed and have RSSFWD email that feed to Twine, which does allow you to email items you want to bookmark. Not sure what went wrong. Will try xFruits next.
- Stephen Francoeur
I still don't understand what Twine will do for me that del.icio.us does not- and the benefit of del.icio.us' large user base is difficult to match.
- David Rothman (☤)
It would be great if Twine did meaning extraction on a web page before you bookmarked it and used results as way to suggest tags (I know delicious suggests tags, but I'd like to see something more robust). See the ClearForest Gnosis addon for Firefox for what I'm talking about http://tinyurl.com/2tjjyl
- Stephen Francoeur
I though this was hilarious. I was ROTFLMAO;) It also showed that the Twine team doesn’t take itself too seriously, something that many of us in our peer/colleague group and in this industry (including myself) are guilty of for sure.
- Josh Dilworth
Yeah the joke was on me today :). I have to say I almost panicked. Until they told me it wasn't the real video! LOL.
- Nova Spivack
Nova, so you know...you will forever be a legend in my books for letting this get out. Honestly this is probably the smartest demo video i've ever seen! Bloody brilliant...the only thing i'm unhappy about is that the tags don't seem to get automatically added to my shit like the shit says.
- Zee.
"this s**t is so dope... you don't have to do a damned thing!" This video is the S**t!
- Jim Hearts FF
You can already do this in several ways. Just add your twine profile page as blog to FF -- this will pull all public items and comments you add to Twine into FF. If it doesn't work you can add the RSS for your profile manually to FF I believe. FF is also going to be integrating Twine at some point.
- Nova Spivack
Let's say you have a large amount of constantly evolving information that you need to 1) easily add to and 2) refer to quickly. What sort of online app would you use?
when you say "constantly evolving", what do you mean? what changes the information? is it changed by you, or outside forces? do the "values" of the information change? or does the shape of the information change? when you say, "refer to the information quickly", do you mean the ability to quickly link/share a single piece of information? or do you mean the ability to query the entire body of information quickly?
- Chris Hollander
Twine's not it, unfortunately. Needs to be more database like. Chris, the info is constantly changing because I'm adding it and tweaking it (this is all text, no numbers). And I mean the ability to query the entire body quickly.
- Carla Thompson
Why can't Twine do that? That is exactly what it does unless I'm misunderstanding the requirements...?
- Nova Spivack
That's exactly what I use a wiki for. Half formed thoughts that evolve over time
- Deepak Singh
many tools claim that they could do it, and indeed they could do partially the goal, but actually none of them can do it satisfactorily well. I guess that's why the question is raised. Be honest, I cannot think of any one that might fully satisfy your request.
- Yihong Ding
Nova - the main sticking point with Twine is that this info needs to be in its own separate silo. When I need to search for something in it, I don't need results from other Twines to show up. It's also a very large amount of information to dump in there. Ideally, I'd love a more intuitive and easier-to-search Quickbase-type application.
- Carla Thompson
i wouldn't use an online app. i'd use a desktop app. i'm a big fan of http://www.circusponies.com/. you can have a notebook-per-silo. you can setup a "clipping service" that lets you select text anywhere on your desktop and clip it in (preserving the source for things from Mail.app and browsers). it automatically generates index pages and has excellent search capabilities.
- peter royal
Hey Nova, I may have to eat my words. I'm just in the beginning stages of trying this out in a private Twine but... I'm liking it so far!
- Carla Thompson
We are doing some work on search features this week and everything will be massively faster in about a week or two (implementing caching).
- Nova Spivack
Try it again. A lot has been added. It makes more sense now. Join some of the top 100 twines.
- Nova Spivack
I just don't understand twine yet either, seems about as useful as LinkedIn. I can output a feed from twine, but I can't insert a feed into it (as far as I can see). Maybe I am wrong about this. FriendFeed seems to "tie things together" much better than Twine.
- Joe....
Is it just me? The interest feed on twine.com's homepage seems random and almost useless (much less than delicious.com-like Zeitgeist). It is just a seemingly unrelated timely list of additions to twine, and it gets even worse in the sense that items which are posted by one user in short time are summed into a completely meaningless posting. Is there a way to get the twine digest on the web? Or do I really have to use RSS as my way of browsing Twine?
- dK
from Mento
It's not random -- it shows whatever is new in your twines. However we are going to launch a better version (more like the Digest) in the October release. Meanwhile RSS works best for me.
- Nova Spivack
You can share them one by one with a public twine (create your own public twine). Alternatively -- in a week or so we will enable you to import your bookmarks from your browser or delicious into any twine -- so you could bulk import your bookmarks from somewhere else into your public twine.
- Nova Spivack
We're more open than before -- you can see our RDF data publicly, but we still have to do some work on what we are putting into our RDF to make it more usable. It's on the list.
- Nova Spivack
University PR groups have a habit of really hyping stuff up don't they?
- Deepak Singh
The sad thing is, the public buys it; they don't seem to care about truth; just about the experience. It this how cultures collapse? They get so wealthy they stop caring about knowledge? Was Azimov wrong and doesn't it end when we stop working, but when we stop knowing?
- Egon Willighagen
Yep this sounds like they basically solved the world's energy crisis -- not sure that is accurate though. I guess we'll have to wait and see.
- Nova Spivack
"In a revolutionary leap that could transform solar power from a marginal, boutique alternative into a mainstream energy source, MIT researchers have overcome a major barrier to large-scale solar power: storing energy for use when the sun doesn't shine." Umm, isn't that what we use batteries for? I'd love to see an actual mass balance to understand just how much more efficient this is than using a battery. Waiting for the journal article to pop up on RSS to read.
- Colin Archer
Ooh, like a lot. Agree with Josh's comment - best way to go about it?
- Carla Thompson
Trying to get the My Items feed doesn't work because it requires username/password to access it (eg, http://www.twine.com/feed...), so I think the only way to go is to use your "My Public Twine" feed.
- Mark Szpakowski
Add the feed from your public twine to your FriendFeed account and it will pull anything you post there into FF. Or, better yet, add the feed from your profile page and it will pull all public items you post in any twines to FriendFeed (however there is a bug with profile feeds right now that will be fixed in about a week -- they aren't pulling the right content. I will remind you to add your profile feeds as soon as it is fixed in about a week!)
- Nova Spivack
same here. it looks, like, powerful and stuff, but i don't get it yet
- Nathan Rein
The idea is that a Twine is a collection of various articles, some submitted by users, and others discovered through semantic analysis and keywords. I may be off a little, but I believe the developers of Twine are active on FriendFeed and can hopefully clear up my description.
- Rob Diana
to me it was too immature when released. i tried to use it twice, wanted to like it, and gave up. i did some comparison of delicious suggested community tags and twine semantic recommendation tags. delicious held it's own so i gave up after that
- rob zand
I use it, and other than the automatic tagging/semantic recommendation features, I'm not finding it all that useful, either.
- cecily
i am going to give the link Nova provided a thorough read and give it another shot - am always interested in finding new sources of info
- Marco(aureliusmaximus)
Here is an example of my frustration. Let's take this entry for example, http://www.twine.com/item.... I found it via my email digest. But there is nothing on this page that is added value from say an interest room in FriendFeed, or monitoring a particular tag on delicious.
Continuing ... my interest is in the "Recommended Items" area, which is blank. The Hakia entry by itself has limited value. Plus what if there is a Twine that I am not subscribed to. I don't want to subscribe to a 100 twines. Ideally, there should be some mechanism that evaluates my interest graph and serves up items of potential interest
- Deepak Singh
Actually that is what we are working on. Also -- as you add more stuff to Twine your recommendations area will show links to things in other twines -- even twines you are not yet a member of.
- Nova Spivack
Nova, that would be great. Looking forward to seeing that
- Deepak Singh
I have been writing recently about a new category of Web apps that I call "Interest Networks" instead of "Social Networks." Interest networks help people keep up with their interests whereas social networks help people keep up with their friends. There's an important difference: While my friends are part of my interests, I have many other kinds of interests such as my profession, products I care about, hobbies, beliefs, places, events, and various other topics of interest. Following specific people is not the most effective way to track specific interests. I need a way to follow the *interests* themselves. That is the purpose of Twine.com (http://www.twine.com), the product that my company is building. Twine is for interest networking -- for keeping up with interests, not just people. Ultimately this is more democratic. In a service like Twine, everyone who shares an interest in topic x is equal. I get the best information from a more diverse set of people -- not just from the most popular people.
- Nova Spivack
from FriendFeed MT Plugin