Live data from Hacker News

HN: Name a problem, any problem, you'd like to see someone solve

news.ycombinator.com

71–80 of 93 posts

Re: HN: Name a problem, any problem, you'd like to see someone solve

#71
The archival of digital photography is unsafe. Photographs that are 100 years old and made out of glass are more resilient that a photo that has been updated to Facebook. I'd like to see a more secure form of digital archiving - better than cloud computing as that is still data and moving parts.

Re: HN: Name a problem, any problem, you'd like to see someone solve

#72
post #38

Here's a problem that I just realized bothers me way too much. I despise the SEO industry. It is polluting the web with useless content that shadows actually useful one. Places like odesk/freelancer.com are bombarded with requests for "article writers" to churn out "articles" with a level of writing lower than the average highschooler. Massive amounts of fake facebook accounts with fake friends, fake twitter follower…

Maybe the new semantic Google update will solve your problem: http://www.seoconsult.com/seo-news/google/algorithm-update-g...

Re: HN: Name a problem, any problem, you'd like to see someone solve

#73
post #43
post #30

Search - I'm even going to pay for it. Search for the damn terms I enter into a search field. The "app" may at most have a CSS style sheet, 1993 look. No Javascript, no images. Give me options to limit my search to date ranges, domains, urls, languages, etc. HTTPS basic auth. No cookies, no tracking, no country detection, no redirect to see what I clicked, no guessing unless I add a flag in the search field. There wa…

It's not EVERYTHING you want, but I started using http://duckduckgo.com/ about a week ago and it's probably the closest you're going to get to a useful search engine nowadays. Full keyboard usage and zero tracking. Simple and clean. I'm enjoying it thus far.

Anonymous seems to like it also: http://thenextweb.com/insider/2012/03/14/duckduckgo-the-defa...

Re: HN: Name a problem, any problem, you'd like to see someone solve

#74
A proactive security tool to protect computers from malware, perhaps using a combo of Artificial Life, Genetic Programming and Artificial Immune systems, with social networking and swarm computing thrown in for good measure.

cf. our current reactive approach of applying antivirus updates post hoc.

Re: HN: Name a problem, any problem, you'd like to see someone solve

#75
post #74

A proactive security tool to protect computers from malware, perhaps using a combo of Artificial Life, Genetic Programming and Artificial Immune systems, with social networking and swarm computing thrown in for good measure. cf. our current reactive approach of applying antivirus updates post hoc.

This sounds both useful and futuristic. However, I think it's more like a new field of research than something you could realize in a startup. We're not even close yet.

Re: HN: Name a problem, any problem, you'd like to see someone solve

#76
1) I would like my car cleaned and refueled while I'm not using it, without my intervention. During the night for example, or while at the gym.

2) I would like a unified api for events that happens and my computer can recognize: pc turn on/off/sleep, received email, a rss feed updated, someone from my contact list is online or goes offline, dusk/dawn, my presence in front of the computer, the presence of someone else in front of the computer, etc...

3) I wrote some scripts to scrape particular image-boards and blogs. I'd like a tool that identifies duplicates and could categorize pictures, maybe with some training (for example meme/macro images should not be too difficult).

4) a music tracker tool (like renoise) with some particularities: notes and durations are not absolute values but relative to a root note, patterns are per-instrument (like in buzz), patterns are zoomable or non-linear in size (some parts have 1 note per measure, others 64 per beat)

5) a file system where I could simply indicate the redundancy I want for a specific file or directory (like a per-file raid).

Re: HN: Name a problem, any problem, you'd like to see someone solve

#79
Music genome project for food. There should be a restaurant genome project where restaraunts and food types are linked in a way similar to how the music genome projects works. That data should be available through an API, which others can incorporate in to their foodie apps.

Speaking of foodie apps, I travel a lot, and while apps like AroundMe are great, they have limited utility when I'm in central Georgia, driving south, and want to know what's along the way. Do I stop now, or do I keep driving to find something better than Chick-fil-a? What I'd really like is an app that knows my route, and the time of day, and recommends restaurants based on where I'll be at lunch/dinner time (give or take an hour).

I'd put considerable effort in to providing feedback to an app that provides that level of utility, even at the expense of loss of privacy. I don't really care if an app knows my dining preferences. I have nothing to lose and a lot to gain.

Re: HN: Name a problem, any problem, you'd like to see someone solve

#80

A software that successfully generates correct Bluebook citations (for lawyers and law students - law students would pay at least $100 each to avoid doing this, there are software's out there, the problem is they don't give correct citations).

This intrigues me, since it looks like some of the offerings are pretty good (e.g. Citrus looks quite powerful).

Would you be willing to chat over email? Mine is in my profile.

Post reply on HN