For "fix advertising," Munch on Me and Groupon were listed. I'm trying to figure out how these two companies have fixed advertising and I just can't. I don't even know why they are mentioned with regards to advertising.

There's still so much room for innovation in the advertising space. Advertiser's think that click through rate is the key metric for them, this is far from true. Click through rate just means the advertisement itself is compelling enough to garner a click. What about after the user clicks? Why are advertisers paying on a CPM basis when they are looking at CTR as the key metric? CTR means advertisers want people to visit a subsequent page and perform additional actions. Companies have tried to assign cost to these additional actions but have mostly been unsuccessful because it is extremely difficult to track. Google Adwords is better for this but people are moving away from standard search (in the long term at least) and it is still far from perfect. Also, there's enormous display inventory out there that needs to be disrupted.

I could go on for a while about advertising and I have some initial ideas for improvements of the online advertising model. I'm excited to talk to like-minded people at the YC ad innovation conference.

I've learned about a fair few local services via Groupon and Living Social. They don't work for all products, but for the products they work for, I suspect they're more effective than advertising. I walked RIGHT BY a restaurant near my office twice a day for 8 months, but I finally tried when it was on Living Social's Instant Lunch. There are a few more restaurants where I work (Judiciary Square/Chinatown in DC) like that. It was more effective than if they had put up posters or billboards and way more effective than if they had gone even bigger with their advertising.

Well, the author of the post isn't totally off the mark.

Groupon and Living Social are lead generation services. They don't do advertising, but they are certainly marketing. The fact that there's even a term that's been coined for what they produce (bacn) says to me that they have indeed had an impact on the marketing/advertising space.

Have they fixed advertising? I'd say no, but i think there are several arguments over definitions/semantics and questions about what constitutes "fixing" involved in assessing that claim.

The new startup DoubleRecall fits under this heading as an attempt to redefine/"fix" advertising.

Checked out double recall, seems interesting but at the same time it seems like a combination of in-line text ads and interstitials. I am sure it will catch people's attention having to type in a product name or words related to a product but I believe that will only annoy potential customers.

I think it will to, depends on your value proposition, I see it working for an article I want to read because the minor inconvenience is worth it for the article. If what your producing is fairly low quality and available everywhere you will drip away a far larger % of users.

I know the creator acknowledges this, but the news sites aren't even close to (or trying to) fixing the news. Maybe someday down the road they will morph, but in fact there has been very little to be done to create an online news organization that both provides original reporting and is substantially better than the web adaptations that traditional media provides. If all we get is better aggregators, the orgnaizations generating the news to be aggregated might still fail.

Personally, I think the best prospect for something awesome to happen in the news sector is that a profitable aggregator similar to HuffPo (but probably not them for a lot of reason), Techmeme or Google News, really makes a run at becoming a media organization. This would require investing in setting up real live bureaus and paying for boots on the ground journalism, but at the rate the current guardians of media move (AP just added links into their stores), it might have some traction.

I agree with the first paragraph of your comment. Another YC-backed startup called NewsTilt was trying to tackle this problem more directly, but failed soon after launching.

http://blog.paulbiggar.com/archive/why-we-shut-newstilt-down...

>aren't even close to (or trying to) fixing the news.

Agreed. I think what is needed is a combination of:

1) Google news (filters duplication)

2) Reddit style voting & subreddits to customize

3) RSS mechanism to access it, with ability to filter keywords

No idea how to earn cash with that though.

Actually the RFS has changed a bit for winter 2012. There are new categories like "New Paths Through Product Space" and "Ephemeralization Apps"

See Link - http://ycombinator.com/rfs.html

I'm not sure if 280 North should be categorized under "web office apps"? Their main products were Atlas and Cappuccino - 280 Slides was more of a demo of what was possible using their tooling?

Cappuccino wasn't really their product either, it was a means to establish a foothold in the framework area... Such that atlas would be popular.

The point 22 doesn't have any samples: A web based database/Excel hybrid. I think Wufoo (a YC company), if used for oneself, could be thought as something like this.

I keep coming back to this idea over and over. DabbleDB was sort of like this but it closed shop after a talent acquisition. Formstack.com (formerly hosted at formspring.com) had very cool table-building tools but it appears they pivoted to become yet another Wufoo/SurveyMonkey. JotForm is somewhere between Wufoo and a database. FormLogix is another form-builder. Hyperoffice.com and Zoho.com/creator sell themselves as MS Access alternatives. FileMaker.com is probably the closest thing to pg's request.

Nevertheless, I still feel there exists a sweet spot that nobody has managed to reach yet. But I don't know how I'd go about finding out if people would really pay for a good excel-db hybrid web-app or not, considering so many alternatives exist, many of them nearly free.

Along with DabbleDB, bList (also acquired, by Socrata) was the other startup mentioned at the time as satisfying this idea.

Also, FileMaker / Bento

I am working on such an app over Google App Engine, which also integrates well with other Google services..

iFreeTools Creator : http://creator.ifreetools.com

Sounds like: http://www.resolversystems.com/products/resolver-server/

Rockmelt: "Simplified Web Browsing" Yeah... right.

How does Gmail qualify as CRM anymore than YahooMail or Hotmail do?

It isn't visible in the 600px version, but the feature of interest is Gmail's People Widget, which is a lot like Rapportive.

I could potentially position my startup under 1, 2, & 7, but since it forces me to pick one I'm skipping it since none of those is likely to be our explicit focus. Story of my life, I've never fit nicely into one box and I'm not gonna start now.

I guess I'll just leave it up to them to make sense of a universal feedback/recommendation engine that's a mashup of Pandora + Delicious + 20 Questions (aka Pandora for Everything).

After looking at that hopefully he asks for easier ways to view,submit and create custom reports from public health data for both doctors and patients, I'll be here Paul.

How is patient's data public? Health is such an interesting space- there is so much that can be done to make people's understanding of their health and their health costs better. The problem is the enormous amount of bureaucracy and the artificial barriers of entry (boards, professional organizations, other semi-public institutions). How do you go about dealing with that?

I don't think I was clear, but my startup allows you to see drug side effects reported to the FDA by physicians, healthcare consumers, lawyers amongst others over the last 6 years. It'll also allows you to narrow down those side effects to age and gender and make custom reports you can discuss with your doctor. The AERS data is public and has been for a long time, but it's increasingly harder to sort. We try to keep the site as clean and easy as possible.

Took me a while to figure out its not RockMe_i_t...domain isn't even registered.

They forgot picplz.

"Asks" for? Paul Graham never asks - the reality just models itself after his writing.