Calendar
Network
Feeds
jasoncalacanis
It was a very personal decision, and I understand that many of you might not understand it. However, while playing in stadiums is a thrill sometimes you feel like playing in your local cafe. I've been getting much deeper, one-on-one relationships with folks thanks to the emal list. 10-100 folks are asking to be signed up a day, and they are making a huge commitment. I'm making one to them as well, so you have this very virtuous cycle happening. It's like the early days of blogging or usenet... intimate, deep, and playful.
The blogosphere is great, but we all know it's become huge and impersonal on many levels. The fact that people race to conclusions with my soundbites (i.e. "seo is a waste of time"), as opposed to actually reading the whole post/listening to the entire keynote, is the perfect example of how blogging has become shoot first, get links second, think third.
Anyway, it's been a great first week.... i'm getting 10-20% response rates on my emails and the feedback is amazing. if you want to be added to the list www.tinyurl.com/jasonslist
OMG OMG OMG... what stunning comments!!! get the torches and burn him at the stake!!! he's saying the obvious...
ha!
@slightlyshadyseo: we're working on not indexing the light content pages. We think any page with less than 400 original words will not be followed. You like that idea? is 400 a good number of words? less? more?
1. fyi, those were sort of experimental pages as I discussed at the time:
http://www.calacanis.com/2007/10/21/best-computer-speakers/
2. we've added some more original content to the page (see the extended Guide note). Do you think this makes the page more worthy of a top 10 ranking?
http://www.mahalo.com/Best_Computer_Speakers
3. We are actually no-following/no indexing lots of our pages that are light on content (i.e. all tag pages and user-generated pages), while producing more and more original content on our pages that are indexed. For example, these pages from the last couple of days are equal parts links and content:
http://mahalo.com/Taiwan_Presidential_Election_2008
http://mahalo.com/Obama_Race_Speech
4. Over time I think you’ll see our pages grow to be over 50% original content, 20% links, and 20% UGC (i.e. reviews, votes, comments). Most pages in the system are 50-70% complete.... over the next two to three years they will reach 80-90% complete thanks to the help of the community and they will be worthy of a top 30-50 ranking in 20-30% of the cases is my guess.
5. We are working on a system right now that will no-follow any pages that have under 400 original words. From what I undertsand from SEOs this will concentrate the Google/Yahoo juice of the site to only the best pages. Anyone have feedback on this concept?all the best,
jason
Someone has got to stop Calacanis.... I mean, really.
Thanks for the feedback.
Agreed that a large % of folks will not care, however the average user is not the only benchmark for corporate behavior and best practices.
Those Twitter accounts are a mashup of our RSS feeds and Twitter designed (and named) for folks who want to follow various feeds. Has nothing to do with SMO/SEO.
So, is this good advice? Will it have any impact on our business? I'm not sure that these little tweaks are so big that they will have any impact on the business... which gets to my point that should I spend our time on these little things or focus on writing another 10,000 great pages?
That's a question... we do all this what does it do? Anything?
I have my positions on SEO and SEM, and I don't think they are the only opinions one could have. I'm more than willing to discuss, debate, and talk about the issues with anyone at any time.
Frankly, the "SEO is BS" quote was a throw away quote I gave at an SES years ago when someone asked me about Engadget ranking so high. I told them it was because Engadget had great content and we didn't focus on SEO--which is the truth. If you have great content and world knows it you'll rank... the SEO/SES stuff is all secondary to having a great product.
That's the INTENT of my statements... of course, at the time I didn't even know the audience was full of SEOs!!! I never knew how many folks even worked in SEO until a year later when Danny brought up the quote for a second time and folks went crazy.
That being said, I've learned over the year that there are many good folks working in SEO in addition to all the slimey guys in bad suits promising the world if you'll give them a $15,000 advance and sign a two year contract. :-)
I'm based in Los Angeles, so if folks want to have a discussion about the issue I'm around...
best j
I really didn't know the SES event was full of SEOs... in fact, after I said it and folks growned I was like "are people in here SEOs?"
I thought the SES event was folks who worked at search engines and advertisers... i never realized that SEO was a such a huge industry. I assumed it was just a couple of black hat firms doing their thing and poluting the indexes.
All this being said, it is going to be a FUN fireside chat and I am going to pull no punches in my assement of the industry.
If you've got an issue with me feel free to take me on at the event!!!
:-)
best j
cute :-)
it's beta software. right now it's "learning" about websites. over time it will be able to look at the keywords it sees across the web and various sites and learn to leave out certain words. While this is happening our dataset at Mahalo is growing. This means our keyword understanding of the page will get better and our data set wider at the same time, which will make the results better and better.
Mahalo for trying to Mahalo Follow!
best j
Never got to meet the ODP/DMOZ folks while I was at AOL (was too busy working on Netscape and Weblogs, Inc.).
However, i can tell you we have DOZENS of DMOZ editors helping build Mahalo right now (and getting paid for doing so... we pay :-).
The links on DMOZ
Happy to see AOL is working on DMOZ again. I wish them well. We looked at the dataset and thought about using it as a starting point for Mahalo since it's GPL. However, when we looked at the linkrot and spam we decided it would actually be easier to start from scratch.
You guys know a lot more about this than I do, but from what i understand it's filled with spam/SEO sites and all kinds of bad stuff. Frankly, I think it's going to be hard to get folks to work for free on DMOZ in this era... good editors are needed on projects that will pay (or donate their fees like we do), so getting editors attention these days is harder.
again, i wish them luck and if they do a good job cleaning it up we would consider using their dataset in some way.
best j
MarahMarie!!! Good to find you here! Keep kicking my butt... you've been a great ombudsman for me for the past couple of years. I always know when I'm close to stepping over the line... you're my guardian angel. :-)
Burgo: some folks just hate to hate... the price you pay for an open, anonymous internet. :-)
DMOZ: In terms of people getting spam into Mahalo, user suggestions are allowed into the user suggested box, but not the main serp. So, it's sort of like delicious (where anyone suggest a link) + and edited directory (like DMOZ) where folks check the links. Our thinking was that anyone should be able to suggest, but those suggestions have to live in the "user recommend box" until we approve or ban them.
Sorry if that is not clear on the site... I thought it was, you're one of the only a few folks I've seen who didn't understand how that works. Users SUGGEST sites to us, then we make the hard editorial decision based on a serious of tests we've developed to evaluate sites (like, are they OK, bad, good, or great... :-).
As i've said my inspiration for Mahalo is: DMOZ + Wikipedia + Google + WeblogsInc + About.com. I've got a lot of respect for DMOZ.... consider this an open invite to any editors at DMOZ to work on Mahalo. Of course, we'll have to pay you--or you'll have to donate the money to Mozilla or Wikipedia.
A bigger question that will come up with DMOZ as an AOL property is should people work for free for a big company like TimeWarner which will benefit from that free labor in a major way. It's the same issue Wikia faces: people are going to work for free to make other folks money?
At Wikipedia it's a non-profit so you don't have the issue (as much).
Robjones: we actually have more folks then we need right now, but I have a rule: ABR.
ALWAYS
BE
RECRUITING
You never know when one of those superstars is going to become available... maybe the best editor on DMOZ is reading this post and has never heard about Mahalo. Maybe they will now email me at jason at mahalo.com. Maybe they just did. :-)
DMOZ: We think folks understand what "top submitted links" means. I don't think we need to spell it out that much. That being said, maybe we do. frankly, the folks writing this immature "i spammed mahalo!" posts are generally trolls and/or blackhat SEOs trying to linkbaint. All of our user testing shows that people get what we're doing and LOVE IT! So, I build the product for the users, not the blackhat SEO bloggers who have it in for me. *
* Note: they have it in for me with good reason: a) I called SEO bullsh@#$t and b) if Mahalo works it's the end of SEO as an industry. There is no algorythm to optimize at Mahalo.... Only humans with really good judgement. :-)
Vanessa: Thanks for the comment...
>> how concerned are you with your target
>> audience being able to find Mahalo?
Great question... I'm not worried at all. We have a half-dozen marketing strategies already lined up, and we've funded the company for a five year build. We're really lucky in that we don't have a clock ticking like other startups typically do. We can really focus on building a great product and let users find it over time.
Also, I've always believed that if you make the best product out there folks will find there way to it. Folks bookmark your site and/or remember the name of it if it's amazing.... so, all you have to do is be amazing. ;-)
>> Also, I know you've addressed this
>> before, but are you worried at all
>> about scalability?
It's a big challenge obviously, but with thousands of people in the Greenhouse and 50/100/200/etc. people inhouse I think we will be able to keep the pages up to date. The public is already checking up on our pages and giving us advice on how to make them better. We only need a couple of passionate users (i.e. site owners, fans, etc) to care about each SERP in order to keep the SERPs up to date.
You can see folks helping out here:
http://www.mahalo.com/Special:MahaloActivity
If you do the math it's not as bad as you might think. If it takes 15-20 minutes to update a page one of our full-time folks could update 25 a day. If you have 100 folks you can update 2,500 a day or 10k plus a month. If our goal is to have 25,000 pages we can update the entire index fairly often... and those numbers don't account for the public helping update the pages. If the public helps and we have 1,000 folks in the Greenhouse we could update the index in a week.
Also, the pages include RSS And other data feeds... so, some items won't need updating. More on this to come.
>> Finally, what about Mahalo do you
>> think really sets it apart from
>> other sites that have tried similar
>> approaches (about.com, etc.)?
We're different than ODP, Wikipedia, and About.com (the three I would say most resemble what we're doing) in a couple of different ways.
However, before we get into that it's important to note that those three sites have all been in the top 100 sites for a large portion of the past 10 years--so having the success of any of those three would be HUGE of course.
We are different than About.com in that we're first and foremost a search experience, not a destination. We're also not involved with the crush of advertising that About.com does. So, I think About.com as successful as it is, has been marginalized because they go nuts with advertising and they really don't focus on the quality of the product enough.
Wikipedia is an encyclopedia, and as such doesn't aim to help you navigate the web/find great sites.
DMOZ was destroyed by neglect--nuf said.
If DMOZ and About.com were really focused on for the past five years I think they would be top 10 sites honestly. So, in some ways their failure is part of our inspiration. I don't think they "failed" because of their value, concept, or the hard work of their teams... I think outside factors killed them/marginalized them.
best j
>> Who do "normal people" go to when they
>> need a new phone, MP3 Player, or gaming
>> system? Their leading edge and bleeding
>> edge "geek" and "gadget" friends and family.
I don't know that you're correct on that one Wolf. While some people do that some of the time, I think a lot of folks do look online for help... I mean, the stats on how many people do car, vacation, college, and product research online are staggering.
Also, it might not be an either or.... some folks might ask their friends then go online for more information I suspect. I know that's what I do.
>> If you can't convince the leading
>> edge, and bleeding edge early adopters
>> that your product is good who is going
>> to recommend it to people further
>>down the line? You'll never get enough
>> momentum or critical mass to reach
>> the tipping point.
Two things:
1. While that might apply to high-tech, risky devices I don't know that think that it's true for all products and services. Not every product starts with early adopters like gadgets or social networks do.
2. That being said, there are a LOT of early adopters showing Mahalo.com to their family and friends. For example, wWe have a ton of early adopters in the Mahalo Greenhouse writing search results and sending their family and friends to the site. We also have dozens of developers making cool products with the Mahalo API and promoting them and we haven't even officially announced the API!).
So, it may not be the case and if it is I think we have it covered. :-)
Frankly, I'm a little surprised at the amount of coverage and debate Mahalo has generated. I thought folks would here the idea and just go "hmmm.... interesting" and then I could go back to work. I'm shocked that in only 90 days we've had so much coverage and debate... in some ways I would have liked to build the product out more. The original plan was to launch with 10,000 or 25,000 pages in 2008--not 4,000 in June of 2007. However, I think it's nice that so many folks are interested enough to debate it, blog it, and participate in it.
I have to say, despite my shaky relationship with the SEO industry they've become the best source or promotion and advice we have! I figured they would ignore us... I guess I figured wrong. :-)
all the best,
J
Danny: submit your link! remember our serps are 60-70% complete when we publish... we're hoping to evolve them with the help of experts and the community!
>> Late 90's dot com people getting a free
>> pass with a different set of rules than
>> people with fresh idesa starting out today.
>> I know of quite a few people with search
>> start up ideas that couldn't get a wooden
>> nickel raised yet you chose to gloat about
>> the fact you have different rules applied
>> to you. My question is do you really
>> need the special rules?
Well, I'm certainly not gloating... the fact is this is a very big project that is going to take years to build. That's why we were able to raise the capital.
Now, five or ten years ago I wouldn't have been able to raise any capital, but since I've done a couple of things that were somewhat successful I've been rewarded with the ability to do bigger projects... that's sort of how the free market works. If you do successful stuff you get more resources in the future... you can see this trend across the entire industry.
I didn't cry about not having resources when I didn't have a track record, and I'm certainly not going to cry about having resources now. :-)
best j
all the best,
jason
Story: Mahalo is BAD for the People
Allen: Wow... great questions.
Many of your questions about message boards, suggested links, etc. can be answered by looking at our activity tracker here:
http://www.mahalo.com/Special:MahaloActivity
Since launch it looks like:
Submitted Links: 3,547
Message Board Posts: 221
You can look at the more popular pages to find those suggested links (like iPhone). If you click on the "Top Submitted Links" box on the right hand side of the page you'll go to a page like these:
http://www.mahalo.com/Special:Mahalouserlinks?page_id=7451
http://www.mahalo.com/Special:Mahalouserlinks?page_id=1822
On those pages we try and answer two questions:
1. Is this a very important link we missed? If yes, move it over to the SERP.
2. Is this a horrible link or spam? If yes, remove it and/or block it from being resubmitted (remember: Mahalo is only for VERY GOOD, EXCELLENT or AMAZING links).
If something doesn't fall into those two categories then we let it sit in the recommend links box until it gets more "votes." (i.e. submissions).
In terms of discussion boards they haven't really taken off yet. I think this is a function of two things:
1. Folks don't know about them/don't expect us to have them (i.e. this doesn't exist anywhere else in the search space).
2. There isn't much to debate yet.
There was one interesting discussion here you might want to highlight in your story:
http://www.mahalo.com/Fluoxetine_(Prozac)/messageboard/
>> -- You had Neil Patel help you in the seo
>> challenge - since you say that "I've only
>> met SEOs who are trying to game the system.",
>> are you suggesting that Neil was trying to
>> game calacanis.com to get you better
>> rankings? If so, please provide some proof
>> as it would be hard for me to believe.
I didn't really have him "help me," he asked to show me first hand what he could do. he did like 20% of what he said he would do (you can check his blog), but we never got to finish the test. It felt like the stuff he was doing wasn't really blackhat (i.e. he didn't make link farms).
>> -- how many of the links provided by
>> "the public" have been accepted and
>> are listed? Examples?
You can look at the tracker and work backwards.
>> -- if a page is about a person, why
>> do you feel the need to repeat that
>> person's name over and over again
>> in the subheaders?
Great question with two answers:
1. We found in user testing that if you didn't do that folks go confused. This was surprising to me, but true.
2. We name our subheads because we will send someone to a subhead if they search for that (i.e. Paris Hilton Photos sends you to hat section... if it was called Photos it would send you there).
>> -- has your position changed on SEO over
>> the past 8 months? Is it no longer the slimy
>> business done by "desperate losers" that you
>> mentioned on your podcast with Fast Company
>> in August? Is Mahalo embracing the SEOs now?
I think there is a portion of the business--a small portion--that do whitehat stuff. I do think the majority of it is blackhat and a waste of time in the longrun.
best j
Story: Mahalo is BAD for the People
Flyingrose,
When folks say "Mahalo is BAD for the people" I think what they are saying is:
"if Mahalo becomes the dominant search experience on the web AND Mahalo's guides are left unchecked AND Mahalo's guides decided to be biased that would be bad for people."
That statement is so wrong on so many levels... let me break it down:
1. No service will be the dominant service--not even Google! There are tons of options out there and people use all different kinds of search engines, blogs, bookmarketing tools, and social networks to find information. It's never going to be "winner take all."
2. Even if it did become "winner take all" it could only be "won" by someone who was user-centric, trusted and without bias. You simply couldn't win a domainat position with a horrible product that was biased--users would leave!
3. Mahalo Guide behavior is 100% transparent and as such any Guide who does anything heavy handed will be exposed quickly, and as I've said over and over again dealt with INSTANTLY AND HARSHLY. In other words, anyone does anything biased or heavy handed they are GONE--as in fired.
4. Mahalo's Guides are all really good people who buy into the Mahalo mission of helping people--I don't see folks becoming "blackhats" any time soon.
So, like many stories in the blogosphere this is a very hypothetical, link-baiting one with a LOT of jumps in logic... the truth is much more boring: we're good people building a good service in a transparent fashion.
That's it.
best j
Story: Jason Calacanis’ First New Email Post (Since Quitting Blogging)