Ask YC: What do you think of our site, bug.gd?

Spend 1 full day researching SEO ranking factors and adjusting your site. Unique titles, meta descrptions, and link names will multiply your organic traffic dramatically. Hugely. If you want to make a business out of this, you should really be SEO experts.

You should also have categories (even if they are derived by keywords in the content), which you should display on your solved solutions page. This will reduce the number of "clicks" a search engine spider has to do to get to the full solution pages.

Bonus, have lateral linking to categories and other problems. So if I'm on a bug for rails, you should have a "view all bugs about rails" link that goes to a list view as well a few links to common rails bugs.

Well, we've studied a lot of SEO (for other sites) and didn't intend really for bug.gd to be that kind of a "content" site initially. I think we need to revisit this, though, so your pointers are quite valid. We got an incredible amount of traffic last year when we were covered on TechCrunch and other sites, but that settles-down after the initial fanfare.

We've prototyped some text classification mechanisms to try to pinpoint the product/category from the error message, but I think we're going to wait until we get more of our embeddable libraries released (that could be included in specific products), our planned Dr. Watson replacement available, and maybe some subsites for particular technologies (e.g. Ruby and Python error messages).

We'll see what we can do, but high on the ROI list would be getting the solution pages less 'deep' for search engine crawling-- thanks for the advice there.

Keep in mind (or FYI), our revenue model is actually not focused on ads or anything of that nature. We're looking to use bug.gd as an engine/toolset to sell to corporations that want to have more a P2P tech support model to reduce help desk costs. Lots of little ideas for this, but this project was really driven by the need for this in a large enterprise I was formerly involved with. The productivity gains can be impressive when your employees have a collective wisdom when it comes to problems.

Due to this path for the company, the core bug.gd site is really intended just to build up solutions to help everyone while remaining free and ad-free. We think of it as a win-win situation and really want all of our new sites to work that way.

Thanks again.

I have nearly 100% success rate solving my problems by searching Google. And if that fails, I usually end up in some IRC channel. What features will make this site better? Maybe have an incentive system for people who solve problems?

I would be curious to know this as well. Any time I see a bug, I feed it (or a search-friendly variation) to Google as well, and my success rate is equally high. And while I enjoy making up fraudulent email addresses for sites that have no business knowing my email, Google removes this step from the loop.

And I agree with everyone else: the design is wince-inducing.

Thanks for the feedback!

We discussed your question in the initial post up above at the top of the page. It's similar to the analogy of throwing everything in the trash versus recycling it.

Nowadays, there's not much incentive for you to recycle your cans beyond kinda just wanting there to be a better tomorrow. Some people don't care at all about doing the right thing and just toss everything into the trash and always will. Others don't mind doing their part (especially if it's a small part).

It's also the argument some people make about Open Source. Why would I release my not-currently-useful source code to others when I can hold onto it for some endeavor in the future?

bug.gd is a lot like that sort of dilemma-- why are you doing all of that error research and losing it to be reperformed by the next fellow? Why does the guy behind you also have to come along, dig through unhelpful forum posts of "solved it myself, thanks", experts exchange hidden pages, etc, etc? If you take the split second to record a sentence or a link that solved your issue, people don't have to repeat your work.

The long-term goal is that you'd find the solution quicker on bug.gd than anywhere else because we specialize solely in error messages and the solutions come with ratings from other visitors in the results.

Also, bug.gd allows you to search against the entire error message, multi-lines and all (with no 32 word restriction like Google). You won't be able to do that on Google, though we've all lived with having to sit there and manually reconstruct our error message in a form Google likes. (Why do that, though?)

In the end, the people who aren't interested in improving things for everyone with a tad more work will stick with Google and be happy. More than a few of them will be taking advantage of people who do want to do their part and use bug.gd.

We're very much in favor of reducing the barriers that make this model less comfortable (e.g. design and email request, see the Firefox extension above that doesn't require that).

For example, we have a feature nearly ready to go live that allows you to simultaneously search in Google and/or an appropriate KB site. The complicated part is helping the user quickly rebuild the error in a form that those search engines will handle properly, but we abolutely want to help them get the solution to their problem however it may be.

In the end it helps everyone when someone uses bug.gd to solve a problem.

Thanks again for the feedback, and I hope I answered your question to some degree.

I wish I could search the database without having to give you my email address.

also: I find the high contrast a little painful to look at.

Yeah, but it's kinda part of the game. Installing the FFE allows you to use the site with an anonymous address.

You can also lie about the email used if you so desire. It's really just there as part of the social contract for using the site.

Any thoughts on a better way to handle that? The whole point of the free/ad-free site is the later reminder to come back and "do the right thing" and help future searchers.

IMHO there is no "it's kinda part of the game". there is good design and there is better design. no opt-in for search should be a no-brainer. remember that this is where you deliver value to the user.

your slogan should be "got error? paste it here."

If they went that route, they would probably lose out on a bunch of answers that people figured out on their own, which is kind of the whole point of the site. Otherwise you have a ton of error messages and problems, and no way to fix them.

hey, what do you think about this story :

1 - the user pastes his error message 2 - he gets a list of possible answers 3 - if he finds a suitable answer, he clicks on it 4 - if he does not find an answer, you ask him to give his email adress so that you notify him when an answer is found 5 - if an answer is found you notify him 6 - after 48hours you send him a message with the subject line "nobody found and answer to your question. did you find one?"

The requirement of an email address makes this a non-starter for me. I think you should be able to search and add without an email address. The reminder feature should be optional.

An OS X dashboard widget would be very useful, as well.

Yeah, we get that a good bit about the email. That's what we use the Firefox extension for, but a dashboard widget would have the same freedom.

The problem is (a) the site is free/ad-free and intends to remain as such, and (b) it's difficult to make that optional because it's really the only reason the site exists. To have some mechanism to remind users to provide solutions (in the timeframe that they remember them) is essential.

That's why the widget/extension model is going to be a big part of our plan, because those can allow the user to get subtle reminders/counts without having to have any email address involved.

Still, we'll continue to brainstorm. Any other suggestions would be very welcome. Thanks a lot for the feedback.

Quick links to save some pasting:

* Site: http://bug.gd

* FireFox extension: https://addons.mozilla.org/en-US/firefox/addon/6138

Thanks again for any feedback you guys have.

The problem with bugs is that the most common ones are already answered and exist decentralized on forums, newsgroups etc, thus making it quite trivial for a centralized placed to take form thus needing a search engine to find the answers.

The other problem is that sometimes people after taking some serious time solving a bug, they are not so willing if not to share without a reward, then comeback to the site and write an answer.

Maybe the colors should be a lot more calming and cheerful.

Nice idea, but to be brutally honest the design is web 2.0 on steriods. How about a small elegent image for your logo and then the form input elements and thats it.

Will do-- light, quiet visual redesign is next. Especially if we get more feedback like that here. Thanks much.

Agreed. There is something about the current design that reminds me of hypno-toad. Although, everyone loves hypno toad.

I agree. In addition to making my teeth hurt, the design wastes a lot of space. Also, it's annoying having to click each heading just to see "link to microsoft KB article"

But it is a good idea. Maybe I'll give a real e-mail address next time.

Yeah, we should include at least some text in the initial solution summaries. I'll see what we can do to get that resolved.

We kinda wanted initially wanted the design to "stand out" and be memorable (i.e. we took a risk) since the most important thing by far, when people stumble upon bug.gd, is that they remember it when they next run into an error message.

Burning our site into your retina was kinda part of that plan, but I think we're over that experiment and will likely move to something that's gentler and similar to my favorite minimalist sites (like YC/reddit/Google).

Thanks very much for taking the time to provide feedback. If you don't want to provide an email, the FFE allows you to search without requiring that-- it's an important part of our "try not to scare people away with email requests" strategy and still keep the site working.

I understand wanting to prevent freeloaders, but it seems to me that most users initially will be freeloaders. They'll want to see that they are getting consistently good info from your site before they start contributing (even so much as contributing their e-mail address or downloading the FFE. And by "most" I mean this is me and I'm assuming I'm relatively normal). Especially when you're competing against every other site full of error messages that shows up in their search results.

It will be very, very, very hard for you to compete with google. If there is a problem, and it is more or less common, there are going to be message boards, IRC logs and various blog entries talking about it. Google finds them all. You, on the other end, rely only on your own DB. Sounds like "one man vs army" kind of deal to me.

Sorry. You asked what I think and this is it.

Google could be their best source of incoming traffic. The various answers sites seem to make a living that way.

Agreed--google will likely provide a bunch of traffic. Also, google is not specialized to this niche. Here's an example from 3 weeks ago where google failed to provide answers that a centralized error lookup site might. Google had little that was relevant for the first several weeks this problem was out and only showed results for an ancient cause of a similar error message. Eventually it picked up a blog post, but 3 weeks is a long time.

http://blog.ra66i.org/archives/informatics/2007/12/23/a-quic...

Quote from the person who "fixed" the google bug with a popular blog post:

There's a "problem" with the current version of rubygems that's causing somewhat of a FAQ, that as far as I can see, Google doesn't index too well...

Yes, I Google for this kind of thing all the time...and quite frequently don't get good answers. I'm a rare bird, I guess, in the sense that I'm usually looking for answers about other people's systems, and so I'm dealing with hundreds of hardware, OS, version, etc. differences. Most folks only ever deal with a few such variants at any given time. Nonetheless, I can see this being a very useful service...not a terribly profitable one (at least in its early stages), but extremely useful. Assuming people use it, of course.

You beat me to it. This was on my "write this for myself or the good of mankind list." The reason it was on that list was I wasn't sure how to make money with it (programmers aren't known for clicking ads). I wish you the best with it and it sounds like you've got an approach for getting money.

I think you're solving a real problem. Google has gotten clumsier about search results, and frequently my searching yields only people with the same question, not an answer. I frequently have to use the cached search result to get google to highlight where, in the mile-long forum page, my keywords appear.

One thing I'd suggest--why use so much screen real estate for each result, but not show an excerpt in that spot? The expandable plus signs being collapsed initially are a negative for me as a user.

Good luck!

I think this is an excellent idea. I like how you prepopulated your DB with msft KB articles.

Like the comments before, let me search and read without giving an email address.

Also, the design is horrible.

Love your domain name.

awesome idea - I've always wanted this site I just wish I could see a list of current and solved bugs (like stories) on the front page instead of having to click, and I'd rather see more content - less logo too

btw how do you solve other people's problems? Your site is essentially a programmer/sysadmin centric version of yahoo answers right? why not organize it that way?

Well, it's a bit different than that. It's really a search engine. People don't (really) solve other people's error messages here. The process is typically (1) search for an error you have and get immediate results, or (2) search for an error you have, don't see an answer, and then come back and let us know later how you got past it so the next guy doesn't have that sort of problem. I.e. stop the cycle of searching once someone has solved a problem.

There is a link in the bottom for reviewing solved bugs, but we didn't expect a lot of interest in that. We may open it up such that others can provide answers for your unsolved errors, but it wasn't really planned to do that.

It's less like a forum and more like a search engine (of search requests).

We'll work on that more content, less logo part. Thanks much for the comments there.

I like the concept, but I'd recommend that you allow people to browse all submitted errors and provide solutions. And, you'll need to give an incentive for people to help. For example, have a list of "Top Problem Solves" or karma points.

The bug in the logo is the reflection?

Technically I guess there's a "bug" in the word "bug.gd", but, yeah, I'm referring to the flipped reflection.

Which has been an interesting dynamic to watch-- there are some people that are immediately hung up on the mixed-up reflection right away and then there are others who think that it's a proper mirror reflection somehow. The comments on our (interesting? bad?) look-and-feel have gotten us at least 10% of our blog references-- we just want people to remember the site and it's been an unexpected way to gain a little extra attention.

change the auto generated user names so people can't tell how many users you have.