Idea Week. Post ideas you don't plan on implementing. | Hacker News Reader

Idea Week. Post ideas you don't plan on implementing.

39 pointsivankirigin107 comments

Having trouble thinking of an idea for a YC application? Thursday looming near? You can take a few of mine.

Each day this week, starting today, I'll post an idea that I'd like to see implemented, but don't have time to work on myself.

Please join me in this empirical study of the value of ideas and post your own!

View on HN

Non-Profit Tobacco.

Find a way to give that majority of smokers, those who don't like what they do (but do it anyway), the ability to purchase high-quality cigarettes whose 'profits' go to cancer research and as-effective-as-possible anti-smoking campaigns.

That actually might work...

I am not so certain. Smoked for almost 11 years before I quit in March of this year.

1. Smokers don't switch brands easily, its more than just choice or taste, bodies get used to a certain cigarette. I wouldn't' just buy your cigarettes because they contribute to a good cause. All cigs are not the same. If I wanted to contribute to cancer I have other ways of doing it.

2. Most smokers actually don't believe that they will get cancer from smoking. 'It can't happen to me' syndrome.

3.The population of smokers is getting younger. An ever larger chunk comes from teenagers and folks in their early 20's who do not want to spend more for cigarettes, additionally, what they smoke is also a function of what their friends smoke.

4. Cigarettes are getting increasingly expensive because of higher government taxes in most nations among other reasons.

5. No tobacco corporation publicly admits that tobacco causes cancer. No one will sell you tobacco to make these cigs,

6. Finally, the assumption that a majority of smokers don't like smoking is wrong. Most smokers enjoy smoking. Ask a smoker about his/her after dinner smoke and that should convince you.

Natural American Spirit helps native americans, right?

"[they] purportedly donate a portion of their revenues to Native American charities." http://en.wikipedia.org/wiki/Natural_American_Spirit

That isn't too far off.

Either way, with taxes, lots of money goes to government mandated campaigns against smoking, tort damages, and miscellaneous government pork. You can't really control that.

Certainly taxes help, but this would target the money that isn't going into taxes -- I think there's still enough profit in the tobacco industry for a project like this to have a shot at working.

Thanks for the interesting link. I like their idea a little, but I see donating some portion of revenues to Native American charities as a far cry from hitting the nail directly on the head with a "real" mission-driven non-profit.

Especially because Natural American Spirit was acquired by a Big Tobacco company.

The tobacco taxes have been misspent. In San Diego none of the tobacco tax money went to health care. Zero, not one percent of it.

Brilliant!

Best idea I've heard in a long time. Better hope the smokers don't sue you, though.

I've always wanted to be able to write notes on top of emails I receive. So say I receive an email about a new bug, I could just jot down more details into another layer on top that email in transparent font/colour. I've noticed other people often print out emails and do something like this with a red marker.

I use Thunderbird and I've thought about digging around in the source to see if this is possible. It's not something I've found time to do though.

That's a pretty nifty idea. I frequently use email as my todo list (though I really shouldn't...I've been trying to do the email zero thing, but I'm back up to 1000 or so messages in my box again, two weeks after starting).

I do use email as my TODO list, but I publish the length of the queue for all to see. Makes me deal with stuff just knowing that anyone could be looking.

http://www.jgc.org/blog/2007/06/measuring-my-inbox-depth.htm...

John.

I didn't know about TB-QuickMove, but have wanted that capability forever. Thanks!

I do just that with mutt. To add notes to an email hit 'e' and it fires up the raw email in emacs (or your favorite editor). Just add comments and save it.

Mutt doesn't have a fancy UI but it is fast. It has no problem handling the 150k+ in my inbox (mostly unread). I know some people that swear by pine and for all the same reasons. YMMV.

Most UI problems are a lack of interoperability. All solved with emacs and flat text.

>Most UI problems are a lack of interoperability.

Most of your UI problems are a lack of interoperability.

In terms of capabilities that users want, I'm right. In terms of making the UI intuitive for the common user, obviously my solution isn't the answer.

Check out Lunarr - I think they do something similar.

If you use thunderbird there is an add-on called Xnote that does just that.

If you're stuck on Windows Outlook lets you edit any message (and you can have what you write be in a different color). It was a neat trick I used a couple of years ago.

Here's one, but if no one does it and it's feasable, some day I might. I'd like to implement a large continual self-report study via email for the public good. Via an email every day or so, data is slowly gathered from participants--all anonymized in a number of ways. The emails would be questions proposed by researchers or lifestyle data to look for correlations with product use, habits, physical traits and disease incidence. For example, looking back in time, thalidomide use and birth defects would be one type of correlation that you might hope to pick up, or perhaps more recently certain shampoos (with ingredient X) might correlate with infertility. Yes you would have to be thorough and careful about the anonymization, but I think the benefits could be there.

Correlation is not causation, though, and the nature of correlating a number of factors dictate you would get many false alarms. However, like a metal detector, the clues for new things to look at would be worth the false alarms. The idea of all this would be to give researchers new targets and areas of research, all free and open.

Thoughts? Impediments?

I like it. Maybe also standardizing the way of phrasing input so as to make it somewhat machine understandable?

Another idea to go along with this that I've thought cool would be to create a wikipedia like system where people logically state their info and arguments in a machine understandable format. The machine would then crunch everything and tell people what the contradictions and possible truths were in the overall system of thought, or particular sub systems.

Combining these two systems could make a very powerful research tool.

Thoughts? The correct analogy is not a metal detector, but a needle in a haystack. I think you misunderestimate the amount of false leads you get doing this by several orders of magnitude. If your system generates too many false leads, it will be (to borrow an analogy from Bruce Schneier, who you should read every week, and writes about false alarms in the context of airport security) like adding more hay to a haystack with a missing needle.

Impediments? People lie about their habits. People do not accurately remember things throughout the day. How to deal with missed days? If people miss days for different reasons, how do you deal with those gaps. You have to assume that the days they miss would be different enough from days they don't miss to invalidate the data of people who miss days.

LPTS, You bring up good points. Self-reports are notoriously unreliable (like eye-witness testimony) however, it is still use-able data. In terms of missing days, the questions can be ordered and repeated to those who miss. In terms of false correlations, you can reduce this by being careful about how you use and analyze the data-set. Multiple regression analysis can be dangerous--it's all in how it is used. If you limit the items you correlate to those for which you have a hypothesis, minimally this data should be valuable in that manner.

I'm not sure whether you are referring to the same argument, but I think I remember that Bruce Schneier talks about the lack of correct positives that makes the application of data mining methods to terrorist incidents useless. He makes the point that very few positives for the system to learn from combined with a very large number of variables causes a huge number of false alarms (not sure if I remember all he said correctly).

So it's all about how many confirmed positives you have and whether you have reliable data on them. It works very well for credit card fraud because there is a sufficiently large number of actual fraud cases, you have reliable data about them and false alarms do not cause major disruption.

I agree with your concern about the reliability and completeness of data. I still think it's an interesting idea if there was a way to extract the data from a reliable source instead of working with what people claim to be the case.

An interesting way to get around the problem of people lying would be to aggregate all the search words a single person searchs and see if there are interesting correlations to other singular points. You could write a plug in that feeds all your searches annonymously to a central database where you could run a gamut of tests.

To your example, when I get a new drug I look up all the side effects (so I would look up thalidomide). Then I have a child and it turns out that my child has birth defects. I would look that up to see who else has that out there has experienced having a child with the same birth defects. As search continues, it would eventually pop up that people who look up side effects of thalidomide also look up birth defects. Then someone could do a serious study on it sooner than later.

As in all cases using data, the major impediment is anonimity and scale of data collection.

A web-based interface for building and managing shaders for 3D apps and renderers. Features would include a node-based shader builder, plugins for the various 3D packages (Maya, Max, Houdini, XSI, etc), suppport for the various shading languages (RSL, VEX, mental ray, Cg, etc), and user groups and permissions. Think sourceforge for 3D artists, small studios without the ability to roll their own management solutions would like this.

Workflow is like this: user creates shader using node editor (think Maya's Hypershade, but Ajaxified...), then sets permissions for the shader. Any user with permission to use the shader can see it from the plugin for his app of choice on his machine and use it. The user sets the renderer that he wants to use, the shader is compiled on the server and sent down the pipe to the user. That way, source code is kept in the hands of the people who created it. Scene renders.

I would love to build something like this, but it requires a lot more CS knowledge than I have. So if you think it's cool, roll with it and invite me to the private beta.

We were going to build a web service to make the college application process easier for students, but don't really have the time right now.

If you like the idea, check out sites like prstats.com or mychances.net (very poorly implemented, but good ideas IMO). Also, let me know and I'd be happy to share our early research with you.

PR Stats...there's a blast from the past.

Umm, I'm doing this right now.

One of my friends working on this, http://www.peerdecision.com/

This isn't quite what we had envisioned. This doesn't really do anything new...

Socratic Blogging. Two or more (real or fake) people have a conversation about a topic and it's posted as a blog entry. I think it'd make for a really interesting new genre of blogs very different from the traditional essay style. A web-based Ajax chat style writing system and an IM client plugin as well, would make it super convenient and painless to create new posts. Traditional bloggers might occasionally do these coop posts together, others might do this type alone.

I had this idea because friends and I sometimes talk on IM about a topic and the chat log turns out to make quite interesting reading.

It would be good for interviews, which are interesting mainly for their semi socratic nature. I posted a similar idea: http://news.ycombinator.com/item?id=65083 http://www.tipjoy.com/our2cents/2007/10/real_liveblogging_li...

I tried to implement this into a blog engine I wrote at my last job, but management didn't get it and directed me to stop working on it. I quit to work on my startup a month later.

I recently suggested a debating platform on Cambrian House: http://www.cambrianhouse.com/idea/idea-promoter/ideas-id/auw...

Very cool!

Create an engine that reads and understands the conversation and turns into a regular article or blog post; notating both authors! Also, create a link to the full IMversation.

Call it www.IMversations.com, if available.

Gosh, I would love to get a shot at YC! Hearing ideas like this is inspiring!

If you like my thoughts & are interested in hearing more ... get in touch :)

My idea is a web service that lets you date Jessica Alba.

Finally a site where a lower user ID is actually important.

Wouldn't that be one that you sell for $50 million?

Depends on how many Jessica Albas your site comes with...

I'm pretty sure it's a severely limited resource.

Not if you use EC2.

The Jessica Alba instantiation feature of EC2 must be new since I last read the docs. What's the API call for that?

I'll send you my instance.

I still think Web 2.0 is missing some kind of app aggregator. Something that would both improve the findability of applications, and allow them to be plugged together similar to what yahoo pipes does with RSS. However, without some kind of standardization amongst the apps, this is way too complex to implement.

And creating a trusted distributed computing system on facebook, where clients would compete to have users give their programs highest priority. On a simple analysis, while this would not be a very dependable system, it seems cheaper than Amazon's service at $1 an hour per core. Plus, this could be made more complex by integrating a contract system - clients can promise a certain payout based on the results of their program, kind of like the lottery.

You mean like diffle.com?

No, not quite, it's missing the pluggable architecture. It would be analogous to yahoo pipes, but apply to general apps, not just rss feeds. It'd be kind of like gluing together stuff in *nix, except in a way the web savvy crowd can.

I think you're right.

Here's one for a FireFox plugin.

The aim is to record articles where you submitted a comment and to be able to track follow-ups.

The plugin has to detect HTTP POST. It could ask the user if he wants to save the page or have an automated process that removes logins and other unrelated actions.

http://www.cocomment.com/ is what you're after I think

does this work with YC?

news.YC has the threads feature which does a similar thing

well, if I understand correcty the cocomment app lets you track comments to threads you are following on multiple sites. i was wondering if it can pull in data from YC and so I can check comments across wordpress, blogger, yc, etc.

An aggregator for social news sites with a recommendation engine to feed you stories you like.

This is something I've wanted for a while too. Interestingly enough, Guy Kawasaki posted this just the other day (he's on their advisory board): http://www.feedhub.com/

Disclaimer: I haven't given it a thorough evaluation yet

I wonder why this needs to be a web service, unless they do the "people who liked x also liked y" thing. Otherwise, I'd rather have this as a local app.

This one has been bugging me too. There's simply too much noise and not enough signal in most of those sites. Feedhub looks like an RSS filtering/recommendation engine. I'm not sure if that's the same thing we're talking about.

I don't want to do it with RSS; I want a web site that organizes the information. I want the recommendation page of Reddit with a better algorithm and stories across digg, reddit and other social news sites. There is probably no reason to limit it to social news sites.

I'm working on something similar at SeekSift http://www.seeksift.com/ though I'm still tweaking the recommendation engine.

Is it legal to scrape from other news sites? Wouldn't that scare off acquirers?

I would start a new reddit, scraping unofficially at first until the bird can fly on its own, and then dropping it.

I'd love to see someone create a virtual lock-box for emails you're tempted to spend too much time re-reading and analyzing (like those from past/future romantic interests). You'd forward the email to another address, which would somehow remove it from your inbox and return it to you after a standard amount of time (two weeks to a month, say).

I would have never guessed that this is a problem. I have a problem with reading hacker.news and other websites too much, and have resorted to editing my etc/hosts file to route those URLs to localhost

you'd want a hosts file for emails? interesting.

The solution isn't to block the distractions. You need to talk yourself up on whatever it is you should be spending time on. If you can't persuade yourself that the other activity is more worthwhile, then maybe it isn't.

i think the idea is more funny than self-helpy. i know scores of women who'd subscribe, if only to banish ex-boyfriends from their inboxes (temporarily).

a tabloid Somebody likes my idea, but for URLs: http://valleywag.com/tech/julia-allison/new-york-nobody-call...

this idea might make women look bad, but i have no problem with that if it also makes some of us rich.

Could this be a regular feature of YC? Even with Google, it's hard to find some of the articles in the past, and this is a valuable thing to keep track of.

Duelnode: Structured debate. http://unqualified-reservations.blogspot.com/2007/10/duelnod...

Just for fun: a graphical lisp editor where instead of () you use div and /div with different color. Something like turing-complete divs. (Then write a lisp tutorial called divintolisp)

Permanent relief from the password hell. And it should NOT be a "keyring solution".

You mean an openID like system that doesn't suck?

It isn't perfect, but I swear by GenPass for this -> http://labs.zarate.org/passwd/

What about a "Windows Update" service that works with 3rd party software. Like apt-get for Windows?

For the software developers who won't change to the new installer, you should at least provide a feature that notifies the user when an update is available and downloads it in the background.

CoMP

Collective Music Patronage

A website that allows listeners to make donations to support musicians. Donations can be directed to specific bands/artists, or to a collective fund, which is allocated to bands/artists based on downloads of free music directed through the site. (The site is a bittorrent tracker.)

The website also promotes communities of music fans, allowing them to suggest new bands/artists to each other, trade news, schedule meet-ups, etc. In theory, the forums could also allow artists to interact with their fans.

The middle-man in music, the record companies, are a Dead Man Walking. CoMP wold be the new, almost invisible, interface between artists and consumers.

Some sort of Web application to actually help with coming up with startup ideas. I knocked at this for a month or two, but couldn't come up with a decent solution. The best I got was some sort of database of user-generated customer pains, which would be ranked (social news style) by users at large. The theory was that once you knew which pain you were targeting, coming up with the solution (second half of the idea) would be easier.

If anyone thinks this problem is even solvable, and wants to work to come up with a good solution, send me an email at keshavs@berkeley.edu - I would actually love to implement something like this.

Hi trekker, if you are interested in pursuing a subset of your idea, I might be able to provide you a canvas for testing it.

We are developing an app that solves a narrow but significant pain point for ecommerce merchants, As we deploy the app to our beta testers, they will all want to add new features. The dilemma we will face is this: which feature should we pick? The answer - and your opportunity - lies in coming up with an algorithm that picks the right feature based on analyzing three factors: (1) the pervasiveness of the pain across the user base, (2) the magnitude of the pain, and (3) our own resource limitations.

Once you apply your algorithm successfully to our app, you should be able to extend it to other apps. If you succeed, this could be boon to the beta community all over the world, and a major success for you.

Thanks prakster. Your comment got me thinking again... a subset like this may be feasible after all. Shoot me an email if you'd like to stay in touch.

hi trekker - I'm part of secure online payment gateway in Asia and we're slowly introducing ecommerce all over Asia - we know about the "pains" of birthing something like this - maybe we can collaborate? email me paulpajo [at] gmail

halfbakery.com

Domestic disaster response (i.e. Katrina II, Bay Quake) using the internet and commercial supply chains to provide sheltering:

http://disastr.org

Lots of room for mapping apps, GIS, route finding, and an enormous capability to handle traffic spikes.

Slashing the risk of financial transactions in the developing world through a free-market inclusive biometric digital ID standard:

http://guptaoptions.com/4.SIAB-ISA.php

Just too big for a start up - they really need BigCo involvement to get going.

aggregating the aggregation sites. Basiclly a site where you can make an account and have the news from digg, reddit, pligg and whatever other sites delivered to this one service. basicly I'm thinking about the user registring on this one site, and then this site registring to all other sites specific users that link back to the main-user. Kinda lika a pyramid ... don't know if this make sense (or if I'm expressing it well in english). This would also be used with social bookmarking sites (apparently 300 of this kind of sites exists - here is the list : http://www.ajaxflakes.com/web-20/social-bookmarking-sites-un...)

And if I'm going there : this site should provide web services that link to user's blogs.

This is my 2 cents.

I'd like to see a web service and plugin for Mozilla Thunderbird to synchronize message filters, junk settings, mail box settings etc. among different Thunderbird instances (running on different computers or on different OSes) - similar to foxmarks for Firefox bookmarks.

Spam attack software: identify the toll-free telephone numbers in spam emails and automatically call multiple times using Jajah (or similar VoIP service).

Unlike spam (which costs little to send), the spammers or their clients have to pay for all those telephone calls.

Same problem as usual with that kind of idea: what stops companies from spamming in the name of their competitors?

A Firefox plug-in like FoxMarks, but for your FF plug-ins and not bookmarks. When you download/install a fresh FF copy, instead of installing all your plug-ins, just install the one in which you sign in and boom it starts downloading and installing your favorite plug-ins.

Combine email, chat, blogging, etc. into one real time system on facebook. Everyone's going to want to start communicating there anyways, since the trust network keeps out spamminess. (I think someone said something similar last time)

Isn't that just a description of Facebook itself? The message system is the e-mail, the Notes is the blogging, and the Wall is the chat...

my point is that they should be one system. Similar to how email and chat are the same in gmail.

http://ikbot.blogspot.com/2007/10/server-side-video-processi... http://news.ycombinator.com/item?id=65320 "Server-side video processing of life-casting robots"

I love iRobot's history of openness. The roomba API is completely open: anyone with a serial cable and some know-how can control it. The ConnectR should eventually be open as well.

I'd like to see this taken to the next level with live-casting from a robot. Justin.tv meets ConnectR.

The robot would either have to be managed by a human when in public, or it could wander alone in a private space. Community voting could control where the robot goes and whhere it directs its gaze. Ideally, the robot would be placed where average people can't be, like back stage at a concert or fashion show. Dangerous places are also ideal. Let the robot go where the humans shouldn't.

Robots with internet enabled cameras can do more than normal robots. Server-side processing means the robot doesn't need to have expensive hardware on board for the intelligence.

Obstacle avoidance, mapping, localization, face & pedestrian detection, object detection, object tracking & motion modeling, etc. can all be done using today's technology with a single camera stream. Some processing would need to be local given the constraints of network bandwidth, but plenty could be offloaded. It is also a good model for premium services: pay more to get faster connections, fast processing, and more capabilities.

Automated surveillance with alerts for intruders could be a killer-app for wifi/webcam robots. Today's choices for home security are immobile, cost thousands, or both.

Server side image processing itself is a viable idea. It's the next step in online video. Today, we just stream compressed pixels. Tomorrow, we'll calculate and stream information about the scene. You can send a flickr image stream to a service which finds faces, builds a corpus of data to identify the people from context, and performs face recognition. Face recognition companies like Animetrics could be tapped to do the hard part.

Motion detection for surveillance applications could all be online, with companies like Intellivid and ObjectVideo already having optimized the image processing component.

Street Views in maps can be combined with GPS tagged digital camera shots to build super high resolution aerial imagery, and eventually 3D.

Services like Fauxto which aim to be photo-shop online could build an interesting api where any image can be sent with instructions for processing.

The point is that all of these services require sending video and images to a server, where some intelligent processing occurs. Often times, the processing will involve the same software modules, so each problem is not unique. Tap the long tail of software development and allow 3rd parties to build their own processing streams that live on your servers. This could be made simple and standard using tools like Python with modules for image processing like OpenCV, PIL, NumPy, and SciPy.

Here's one I'm dying to implement: user-generated reporting website for the wait times on the rides at Disneyland. I swear Disney distorts these for easier crowd control.

If you stay in one of the Disney-owned hotels, you get to skip the lines on all the rides. It's a little more expensive, but it's worth it, since you wait five minutes, tops, for any ride.

No, if you stay at a Universal owned hotel you get to skip the lines on all rides at Universal theme parks as many times as you want. Outside guests can buy this feature for around $40 a day but you are only allowed to skip the line in each ride once.

Disney's virtual queuing system, called Fastpass, does not create different classes of users. The only people that get special privileges are those that are doing something like Make a Wish or winners of a contest and these extremely limited superpasses don't impact overall wait times. The advantage given to Disney guests is that on each day a park will be open one hour early and up to three hours late. This isn't that much of a benefit during peak times because the park that was open early will have more guests than the other parks and you have to get up an hour early for optimal touring. The late hours aren't that much of an advantage because everyone wants to stay late in the park.

A naive perspective on Disney's Fastpass is that it doesn't actually impact total waiting time because the standby lines will take twice as long. This is incorrect because Fastpass lets the Disney engineers redistribute traffic however they want, which measurably reduced wait times for the most popular rides right after the implementation of Fastpass. Wait times are still higher than they've ever been though, because park attendance keeps going up. Disney is a profit machine.

For more information about waiting in line at Disneyworld, I strongly recommend Industrial Engineer Bob Sehlinger's The Unofficial Guide to Walt Disney World, whatever the latest edition is. He also mapped out a mathematically near-optimal plan for riding every single ride at the Magic Kingdom in one day. I really want to try doing that one day.

You're right. Funny how the Florida theme parks of one's youth tend to blend together.

Slate has a good (though somewhat old) explanation of both Universal and Disney's systems (http://www.slate.com/id/2067672/sidebar/2067676/). If its numbers are still more or less right, using Fastpass reduces your wait time to about 15 minutes when you come back to the ride during the window you've been assigned.

They do distort them for easier crowd control, but they only overinflate wait times.

I realize some misdirection on my part. I should post my ideas in this context too. This thread will be gone by Friday, so I'll continue to submit ideas as separate posts like these: http://news.ycombinator.com/item?id=65320 http://news.ycombinator.com/item?id=65083 -------------------------------------------------------------------------

http://www.tipjoy.com/our2cents/2007/10/real_liveblogging_li... "Real live-blogging, live comments, and a comment API"

Live blogging is an interesting aspect of the blogosphere. Bloggers post in real-time, often while attending an event, chiming in live with their thoughts and views. This is a great way for bloggers to communicate with their readers, but it's currently not well supported in blogging tools.

Live bloggers should be able to update their posts in some manner similar to a chat client, or collaborative editing software. The system would take the blogger's new content and appended to the post on the readers screen without a page refresh. It would be entertaining to watch something so live, along with a different take on the process of posting to a blog. Readers could see sentences edited and thoughts refined.

Ideas come from writing them down, and this would be a live crystallization of thought.

The stream could be synchronized to an event, and then later could be played back to get the same experience.

Similarly, comment sections could be live. This is nothing new, with group chat rooms updating messages to the whole group. But it would improve the user experience of a comment thread, with new comments appearing on the site live. For sites with comment modding, the votes could also be live -- watch trolling comments nose dive and disappear.

Regarding comments - comment systems should have open APIs. For example, if a blog post makes it to the front page of reddit, digg, and Hacker News, there will be 4 disparate comment threads, one at each of the news sites, and one at the original blog. Rather than forking the conversation, recommendation sites augment it: grab from and push to a comment feed. Comments on Hacker News should be visible on the original blog, and comments there should appear on Hacker News. (Except YouTube comments; no one needs to see those.)

I already mentioned a few of these points to the Disqus folks. They certainly have a good system on their hands, and any real competitor would have to do everything they do, and maybe add a few of the ideas here.

"... Live blogging is an interesting aspect of the blogosphere. Bloggers post in real-time, often while attending an event, chiming in live with their thoughts and views. This is a great way for bloggers to communicate with their readers, but it's currently not well supported in blogging tools. ..."

One start would be to create an opensource wordpress plugin + a client side browser bit of js code to support it. That way each user could support their own comment systems for each place they comment. The bit that currently alludes me is the discovery bit. How do others know you have commented on a particular site?

One way could be to publish say RSS feeds that link to sites you have commented on allowing others to discover what others have said. That or some sort of microformat that marks up some page that people read. This would not be real-time though. Commercial sites have an advantage here as they have a centralised place people can look.

Maybe the solution is to offer a commercial discovery site that acts as a registry of comments. So for example:

- user John writing using open source firefox plugin you make some comments on a third party site say hackernews

- plugin sends comments back to your open source wordpress blog, adds comments to blog then generates new RSS feed of comment detailing comments + discovery information (blog, url etc).

- wordpress pings a commercial site "CommentsRUs.com" that pulls the RSS file, parses it and either creates a new entry for a particular site, adds comment.

- user Jane who has subscribed to CommentsRUs.com, HackerNews site is notified by twitter that John has responded to a HackerNews article she has also commented on

- user Jane makes another comment and CommentsRUs.com either notifies user John or makes an entry on his homepage http://commentsrus.com/person/john/hackernews/ so he can follow the thread.

Of course this is no substitute for the original site because the annoying thing is the meta data is not captured. Then again if they did have some mechanism to capture the metadata ( http://goonmail.customer.netspace.net.au/hackerid ) you can do lots of useful things.

teleport

a site where you can use GIMP 2.0 plug-ins like fake lomos, holga effects etc. Fotoflexer does this using it's Warholize effect but it's limited to that. Maybe the site can have apps/plugins integrated into it by user ala-Facebook so that you can have a library of effects after a few months built by the users

There's a few online GIMP/Photoshop type apps already. Most are done in flash and work pretty well.

One of their advantages is that plugins are (of course) run on their servers, and thus even low powered desktops could have their filters be applied instantly.

So what happens if someone suggests an idea but changes his/her mind later and wants to implement it after all?

Perhaps it would be better to just ask people for ideas that they are willing to share -- without any implication/suggestion as to whether they plan to implement it?

They are forced to build a better, more useful, competitive product so that they, their investors, and customers all benefit.

I mean the people who used the suggestion would be upset because it seems that the person suggesting it said he would not pursue it.

Valid point.

Isn't the point that almost everything is in the execution?

That would make the guy in question an idiot.

Why? Discussion is supposed to change one's mind about its subject.

You discuss ideas with partners, not with potential competitors :)

Reply on news.ycombinator.com