I've been thinking about the problem of making make people organize their public and semi-public data better.
Groups (such as email groups and Facebook groups) and communities need a way to organize and share their data freely, but in a structured way. Also people who have data, or start building some kind of data, in text format, or tabular data, but don't know of any other mean, so they build a blog and start posting everything, then it is impossible to make anything useful with the data later. Or maybe they make a spreadsheet and link to it.
I've come to 2 solutions:
- for the first thing (but also applicable to the second), a structurable database (such as PouchDB) stored in the people's browser and communicating with other people from the group via webRTC could build a decentralized database everyone will have. Some things could be made on top of this;
- other thing is just a blogging platform that is not only a blogging platform (what I really want is it to not be a blogging platform), but it is also some way to publish static atemporal articles, tree-data, tabular data, hierarchical data etc. Written in javascript, running from the browser and communicating directly with GitHub or some other static hosting service with an API. The person would load the contents with the API then deploy the new contents as a static page. Free blogging, no central servers or providers (except for GitHub, but it could be made to generate HTML hostable anywhere), cool stuff can be attached (even some plugin that will take the data from the first solution and share in the blog/page/site).
Immediate sharing of raw data (with the world, or restricted to one’s peers), instead of going through the tedious process of information packaging through text, words, written description or synoptical graphics — that’s the future of our open Web. When anyone can create and update the underlying raw data, there can exist infinite summarizing views, syntheses, charts, text, all drawing their own conclusions.
Pushing and syncing strings is a solved problem, as are atomicity, consistency, isolation, durability. Powerful visualization tools are being made, too. The View and the Controller are taken care of.
The unsolved problem remains: how will we empower users to relate those strings, to have them turn data into information, and information into knowledge? How will we enable the many thousands of non-tech domain experts to create their models? How will we devise the Model?
We’ve working on an approach to tackle this problem. We saw that the burden of having to create a data structure - along with the need to attach it to a visualization - makes people post what they can as hacky html lists, if they get so far.
We wanted to show people that they were in fact able to create structured contents, even if it was in a limited way, and we choose the analogy of file and folders since most people know how to organize them. But structuring and gathering data is half of the problem, you need to translate that data into friendly views and same sort of browsing experience, so we developed same conventions to turn the structure into a small website/embed.
We thought of developing an open way to store and share this data, and let people copy, follow, contribute to these chunks of content. Build a friendly web dashboard that exposes every item with simple views, and which let other developers consume them in any other ways they want.
The problem is that we don’t really know who wants this. We thought that it would be appealing for web agencies as a way to quickly add structured contents to websites, but we were wrong. Some people suggested us to develop a full CMS, but we wanted to work on the problem of sharing and opening up a library of contents, not to be an alternative to Wordpress. We honestly don’t know what to do next. If you want to talk further about it, my email if fsainz/gmail.
If you're talking about https://www.pagebox.es I see it as a good competitor to Trello (I always thought of using Trello's API to build a simple idea-tree note-taking app, but their trees have depth limits, yours don't, so I could use yours), but, at the same time, I don't know if your site will be up and online for a long time, so I'm not sure if it is secure to put all my data there.
We've never compared ourselves to Trello, it makes me think that we could make the nestable lists more explicit and promote this note-taking use. As for how long we will last, it's always hard to say, but we won't take down the server, and in any case be assured that before that I'll help everyone to take every byte safely elsewhere, via zip, sql dump, csv, spreadsheet, plain html + css, written in clay tablets.. I mean, I'll be your export module, If you need me to programmatically push it to other service with an API I'll do it, trapping the contents of someone else makes me feel terrible.
Now I'm thinking you could explore more the multi-useness of it. First, you expose an API for accessing the nestable lists from the browser (and other clients, but I like the browser), then various apps can be built on top of it, and they all could be accessed from the main site, such as:
- note taking
- server logs
- manual logs of events (I've seen these in use in lots of small business where some employee must log what happened in its turn)
- personal journal
- mixpanel-like thing
- analytics of any kind
- computer-generated logs of anything that doesn't need a full-fledgeded database and system of its own (I'm thinking specifically of the logs of people who used their password to open a dor we have in our building, that communicates via serial port with a server that stores that info, but very ineffienciently)
I wouldn't know what to do with Pageboxes also (as I still don't know of what exactly I am talking about in the main comment in this thread), but my use cases for the general idea were two:
- Online communities. For example, groups of students of some area, groups of pregnant women, groups of people who like tiny houses, groups of people who do geocaching, people who like to save dogs in the streets and give them to adoption, people who like to shoot at birds, people who work with something specific. All these groups probably have groups of email, or Facebook groups, (probably more than one group, maybe various localized groups) where they chat and share interesting links, data and information. Then this data is lost, it isn't searchable by the people who know it exists (specially if they are inside an Excel spreadsheet that was emailed to all the groups' participants), it isn't knowable for people who joined the community later. This people will never know (or will know too later) that that interesting data was there, just behind a pile of emails and posts. There are a lot of examples of data that could be indexed (yes, I'm talking about database indexes) in interesting ways, but I will give none now, and lots of data that could just be organized in an easy way (categories, although I don't like this very much) and made searchable.
For this problem, Trello is a good option, but it is limited, it is centralized, it requires an "owner" of the Trello board to authorize other people, it cannot develop in an organic way, with few people starting to use, then other people joining, and people who don't wanna join never joining. It also cannot index data, just categorize and archive.
- Myself. I don't want a blog, but I have interesting data stored in my computer and my brain I wanna handle to the public over the internet, to whoever be interest in it. I have articles I can't find anywhere else anymore, I have spreadsheet data that should be presented as spreadsheets, I have articles I typed from old magazines, I have videos I don't remember how I got, I also have impressions of books, commentaries about lots of non-mainstream topics, and mainstream topics also, mainly atemporal things, that don't fit well in a blog. I could make a static web page and put all this there, but it seems too difficult for such a low value it will have. I also don't wanna do it because I wanna keep thinking about how would a non-programmer do it. I can't help thinking every people in my situation would: (i) do nothing; or (ii) make a blog, post everything there for two month and abandon the blog. I don't like abandoned blogs.
Pageboxes could do something for this, but I don't know how exactly.
Long time ago, I mused¹ about these kind of things, then worked like mad on devising an implementation. Meanwhile, a side project I’ll take up again, one day.
I love the idea of data blogging.
Thinking more along the lines of what Freebase was doing (before Google smothered it), what Fluidinfo² originally intended to do, and what Silk³ seems to be doing best, right now.
"data blogging" sounds awesome, never thought of those words. I'm pretty much addicted to RSS Feeds, for me there's no better way to access those contents. Being able to translate that experience to structured data would make very happy.
Reading your tumblr post, I think there is useful work to do on the points you mention:
Collecting and List Making // Visualising Knowledge // Open Access. We need something simple enough to get the people who already know how to publish posts interested.
Silk is pretty good, and I never heard of Freebase. From wikipedia: "Freebase is a large collaborative knowledge base consisting of metadata composed mainly by its community members. It is an online collection of structured data harvested from many sources". I got to look into this, sounds beautiful.
Yes, I was talking about Pagebox. The current version is more like a page builder with some structured data, but we've always dreamed of turning it into a "github of content data", if we can call it that. We are thinking of changing the interface and emphasize the ability to navigate and search, possible something like a mac finder / itunes style of dashboard.
I love the idea of helping build these community databases and make those contents searchable. I see a challange in the fact that this means adding another channel to discuss contents different than the one they already use, but the promise of making those contents easier to reach might make it worth it.
As for the blog problem, I'm in a similar situation. A blog seems to have some lifespan, and there are some raw contents that I would like to have available without worring about a domain name and the design decisions of CMS theme. I have my own particular "delicious", some comments about different articles, a bunch of photos, and many screenshots and mockups from the different projects I've been working on that I would like just to publish without fuzz (Ideally: add all my project folders. Done). It won't be as beautiful as those web portfolios but I don't need that, I want convenience, I don't like abondend blogs either and I have 3 or 4...
Reading these comments gets me really motivated, thanks!
Analytics on all of my application usage, email, chats, texts, twitter and facebook interactions, and map locations. Should provide outputs like Google Now, but run on my own operating system (preferably Linux-based).
This (http://www.jwz.org/blog/2014/01/psa-back-up-your-shit/) should be good jumping off point to get a lot of the stuff. The system should have customisable hooks written in something like Python to write your own "filters". e.g. "If I've been in emacs for 1 hour straight, pop up a reminder to take an eye break, etc."
RescueTime [1] seems to be along the right lines here, and is available on Linux. Doesn't appear to be able to track emails/texts/social interaction, though, only the time spent in applications and websites.
Now that the smartphone innovation has slowed down a little, why not bring back real buttons, wheels etc.? They should be well-designed, precisely engineered and fun to use. I belive mobile gadget design is now awesome enough to make this work.
I wouldn't mind a scrollwheel, or even real radiobuttons (one button goes in, the other pops up). The exact functionality would still depend on the current application.
Since they give real physical feedback they should objectively be superior to the touchscreen-only functions, no?
A browser extension / additional service that allows you to tag/lookup the music on google music / spotify by any set of strings. And I mean your own saved collection, not for sharing/exploring/social purpose. This can be faked to some extent by multiple playlists, but it's a bit hard to use in real life.
This would be great help when trying to dj dancing party: now show me a list of "blues", "chicago", "slow", "with breaks" to choose the next song. Actually in this context, separate service / app would be great - it would work on a mobile device, rather than just desktop.
About to get started writing a little script that will take care of synchronising my library with my google music one, and automatically uploading those that it can't find a match for. A few of my friends also use it, so I was also thinking of making a little tool that would allow us to compare our libraries and show us what each other are missing etc.
A mobile application to assist in the job of house/flat hunting.
Something that could tie in to RightMove (at least in the UK), access your saved properties, and provide a simple interface to schedule viewings (although I guess this would still have to be a manual process to book viewings, but the app should provide a calendar to store them in).
The app could then alert you when the viewing is soon, and provide an interface for you to rate the viewing, add comments and maybe your own photographs to remind you later on, rather than having to scribble notes down on the same piece of paper that becomes home to your chewing gum the next day.
On a social front, it would be good to see how many other people were viewing the same property, to gauge the interest, and maybe you would be able to see an overall rating of a property based on what other people that have viewed think about it?
House/Flat adverts are still relatively useless in the process of finding somewhere suitable, and until you can view and see a property for yourself, it's impossible to tell. I think it would be very useful to be able to see ratings and comments from other users, such as "the second bedroom is a lot smaller than the photos show, and the flooring in the bathroom seriously needs replacing!".
It doesn't look like there's an available API for RightMove at the moment, so I'm not sure how the integration could happen. Still, a stand-a-lone app could still be a worthwhile idea?
The rental market in London is pretty ferocious. I don't think comments could work because you'd always run the risk that somebody has decided they like a place and want out off other people from looking. In the end you'd still have to check the place yourself. I like a lot of your other ideas though and user photos could definitely work.
I agree with the user comments issue. Maybe a way of vetting them, possibly if other viewers agree with previous comments, they become publicly available?
EDIT: ...but even then, you might not get personal opinions. Tough one!
I agree it would be nice to bring renting (in London) etc into the 21st century but you're unlikely to ever incorporate a yelp type ratings system for property as there's no incentive for landlords / agents. They rely too heavily on spin and sales techniques - they don't want you talking to other people and discounting properties before you've seen them. They want to take you on a journey where they lower your expectations so that the one property they show you thats better than the others but slightly out your price range makes you fork out the extra cash. Estate agents and landlords want to control the market - you'll have a hard time democratising it. It really is a sellers market in London.
I'd personally like to see the use of web/tech to integrate the whole experience - so everything from deposits, rent, grievances and so on is centrally managed and moderated. Again though, hard to get traction.
A public protocol and a reference implementation of a social network. The problem is stabilized, it's time to do to social networks what SMTP did to email. Open it up to be fully decentralized, where people can host their social hub or use a hosting service, of which there would spring up a multitude.
Social networking is nowadays as central to the internet as email, and it is the first such, heavily used, application layer "protocol" that is closed.
For a small period of time I had hopes that this would be the path G+ would take to the top. It isn't.
Diaspora does offer decentralization. A node servicing a social network can be hosted almost anywhere. I was very excited about that project early (and it still has potential) but ultimately I don't believe its backend oriented advantages are going to woo people away from centralization. In other words, to average users it looks very much like a FaceBook knockoff.
I believe integrating social networking into email might be the most promising approach. I've always thought a "ShareBox" icon below my Inbox icon would allow social networking to decentralize in a way that would appeal to end users.
Shameless plug for a research project going on at my school: https://priv.io/ can replace most social networks you can think of in a decentralized, private, encrypted manner.
A progress and time tracker like Habit RPG and Rescue Time that tracks number of hours spent towards a habit/skill. Basically, something that can track where you are on 10,000 hours rule. More automation is better!
That doesn't allow 2 windows comfortably though which is what shoots dev productivity through the roof. Higher res big screen 16:9 would be better, but then you reach the weight limit.
How about a search engine that bases its results on how unique the information is on every one of them.
Every time I try to do an in-depth research on Google about some topic I open up 10 or 20 of the most relevant results, only to discover that at least 80% of the information is repeating itself in at least 50-60% of the pages.
One could write a search engine that compares the contents, groups the articles that have similar enough contents and only displays one article per group. This could work especially well in the case of longer searches with enough keywords.
Obviously, this is not a standalone product that would compete against the established search engines, but it could be built on top of them - a browser extension for example.
The problem is lack of semantics. Computers don't understand the content of a web page or article, and you have to manually parse it and make sense of it. You'll find 80% of what you need in 20% of relevant pages, and the remaining 20% in the other 80% of pages.
The semantic web is the solution. A system that understands content, understands what you know, and understands what you want. Then, all it has to do is show you what's relevant and necessary to the task you're trying to accomplish. In most cases, you won't have to actually do the task, as it will be automated.
I don't see any way around this. Changing how search engines work will still not let you find articles that complement each-other perfectly. And you'll still have to read more than necessary.
Yes, the problem won't be solved completely and you will still end up reading the same content more than once. But at least, by analysing the text the search engine could detect which articles are coming from the same source (e.g. inspired by the same press release or just telling the same news event over and over again) and group them accordingly. Currently, I don't feel like Google or other search engines are doing any of that (maybe they are, but if so, the results are not good enough).
A bookmarking service, that not only saves any "interesting" links that you visit but also saves the data in that link for future search. The service should let you specify topics that you are interested in and will analyze every page you visit (with the exception of blacklisted pages) and put it into the appropriate bucket (topic), for future reference. It should also index the PDFs and word documents that are read by you. In research we consume so much of content. Organizing them effectively becomes as daunting as the research itself at times.
>> The service should let you specify topics that you are interested in and will analyze every page you visit (with the exception of blacklisted pages) and put it into the appropriate bucket (topic), for future reference.
Try ToRead.cc. It will mail you the contents of the page you are at, and allows you to add text to the mail that will be sent. I use filters and topic tags to classify any interesting links I find.
Not easier. You have to wait for the OS to load, login, find the app you want in your menu bar, etc. I think Wine just opens the app in you current desktop environment and it's presumably faster since it doesn't need to emulate the CPU.
Ah, I thought they did (at least the CPU architecture, not down to the actual transistors/atoms). Wouldn't a full blown VM which also needs to run the OS have more over head than Wine though?
They emulate the CPU if you're using a different architecture, that's why android emulator can be slow. You can just "hibernate" and have the application open in the VM at all times whenever you turn it on.
It's not easier by a long shot; for multimedia programs that deal with large files it's always complex to set up "shared folders" or whatever they're called; and you run a whole emulated machine just for one program.
You should take a look at VirtualBox's API. I have a couple of engineering programs that I need Windows to run. I have entries for them in my launcher which start up my Windows VM in seamless mode and launch the desired application. Apart from the different window decorations, it fits perfectly into my standard workflow.
Ah ok thanks, so yes, CrossOver would be what I'm looking for. LightRoom is mentioned as "known not to run" so that's a bummer, but Vegas 11 has a Gold medal for Linux, which is a good start!!
I could certainly use a better system for remembering where I park my bike.
One system: Take a photo of the bike every time you park it in a non-standard place (standard being e.g. "work", "home" or "girlfriend's"). Unfortunately, taking out my smart-phone every time I go grocery-shopping is bothersome, and I might only find out I need to suddenly go elsewhere when I've already parked the bike.
Ideally, my bike (or, through statistical inference, my smartphone) should just publish GPS coordinates. Unfortunately this drains the battery very fast. Perhaps a background service that uses the GPS a little more sparingly than the Google Maps app?
A media browser for leveraging (browsing, and surfing) multi-terrabyte home media servers:
Problem: I've got 2TB of videos, music, and images on my hard drives, and I've found my available tools to be too low-leverage on managing this anymore.
Context: about 2 years ago, I've stopped the directory structure madness, and embraced search as first-order link for media files. As a byproduct of this, I notice I don't have a clear mental structure of what videos/music is downloaded to my computer anymore, and to where (good riddance!). Neither do I want it to; instead, I'd like to have a "youtube for local hard drive":
Basically, I'm looking for a local Media Browser application for windows: specifically, an app which allows:
* thumbnailed preview of all videos & images
* Automagically categorized based on all online, or within-file available meta-information
* With a built-in search
To perform the following operations:
* Preview a thumbnail on video files (similar to file explorer)
* Display categorized lists of all media available locally on my drives
* Channelsurf, similar to youtube: be able to start from one movie file, and have "similar" videos around it
* Double-clicking on the file to start mplayer
* typing the first 2-4 letters of the movie/image to instantly bring up the list of media with that part in it
While specifically not doing:
* Anything that requires me to touch any of the files in any way shape or form whatsoever
* Any upsale (I'm looking at you, windows 8 "personal videos"), advertising, or introduce any friction during interaction
* Bringing up any browsers: solution must be windows-native, and responsive (<20ms results)
Specific things I've tried so far:
* The closest thing I can wave towards "want" is apple's itunes, and JRiver's media center; they both share the same weakness of not being able to use mplayer as playback engine
* Banshee on windows is unable to show video previews, have multiple stability issues, and can't use mplayer for double-click
+1 for Plex. I almost gave up on the idea of a HTPC after the headache of correcting hundreds of mislabeled files in XBMC. Since I switched to Plex, the problem has all but disappeared (I can think of 3 or 4 things I've had to manually correct).
Plex shares the terminal disability (with itunes, and JRiver) of trying to be two things at once: a media browser, and a media player; quite similar to how file sync programs before dropbox were terminally unusable by trying to include redundant functionality ( http://qr.ae/rwZF4 ).
Regarding built-in media players, not one vendor takes usability of playback seriously. Specifically, when I hit "enter" on a file, that means I'd like to see it played back, like, now. Not "queueing", nor 5 seconds later. When I scroll left & right, that means I'd like it to go +-1 min / hour, like, now; and not 5-10 seconds "seeking". On my quadcore 3.4ghz battlestation, this remains a problem for every single native, and web-based app. MPlayer is, to date, the only one which consistently delivers on sub-second videoplayback, and it does so from cold start.
(There are also major problems with how subtitles are handled by players, but I grant you that might be a niche)
Hence my strong requirement above for the app to go with shell.exec, instead of half-baked non-implementation of every single video codec under the sun.
I started on something like this here, https://github.com/wallace/zuggy. I imagine a sqlite3 db with a rails app with vlc in a docker container. One day I'll have the time to make it happen but for now, there's too many other projects on my plate.
edit: Also, I'm not sure how to make it a viable business.
The wife approval factor of Windows software is zero. This won't fly in the living room.
I want this on a set top box... And anything like yamj or xbmc absolutely chokes on my collection and I spend hours fixing incorrectly identified files.
started doing something with this with my automation system https://github.com/dandroid88/webmote. endedup just leveraging mpd instead but when i was working on it it was much like you described where you could walk the server/discs and just play music from the browser screen. From there I planned on all the meta data enhancments but then got busy with other things. Ill find the computer with that branch and push if i can.
Email service for small business: something companies can use in combination with Box.com / Dropbox.com or other cloud storage instead of Google Apps.
This is not really idea but more of our need: we started using more Box (and Dropbox) instead of Google Drive but we are stuck with email provided via Google Apps. There is no other service on the market as Google Apps email for small business.
(I tried FastMail, Atmail, Rackspace, and none of them is a good alternative).
- does not integrate with third-party systems (yea "not their problem" but they should do something to promote 3rd party system to integrate with them)
- spam filter is not so great (false positive is the problem - I'm ok having occasional spam in inbox, but email from wife linking some stuff I need order is really annoying.)
I'm pretty sure that it is not what you are looking for, but http://proxy4.us can forward your domain's emails to your existing inbox (like @gmail.com).
Oauth for contact details. When I change my phone number or address, the individuals or corporations I have authenticated should be able to fetch that latest info from a centralized service that manages only that information (and holds it in strictest confidence, no list-selling).
A completely transparent social platform with messaging, activity tracking, content management, knowledge storage, wishlist/preference/opinion management, idea generation, etc.
You might not know yet, but privacy is the root of all evil, and the privacy madness has to stop.
The more we hide, the more we have to hide. It's a never ending cycle.
People don't see the world as it really is. They see the world the way others want it to be seen. They see an artificial, curated version of the world where people have no flaws.
This disconnection between what is real and what we think is real makes people feel broken, different, not good enough. How are they supposed to know that nobody is perfect when everybody actively try to show their perfect side? We all wear masks by fear of rejection, fear of being different.
Just like you, I need privacy because privacy is expected of me. I need to lie because people don't want to hear the truth. I have to wear a mask because everybody else does. Being completely transparent and honest is a sure way to be rejected and jailed. But if I had the choice, I would prefer to be myself, to never have to remember lies, to never have to hide.
Do you think the privacy paradigm is sustainable? Do you think it makes sense to use so much resources to encrypt everything? Do you think it makes sense to hide yourself behind proxies and firewalls? Do you think it makes sense to protect everything with passwords? Do you think it makes sense to keep creating fake profiles online? Do you think it makes sense to disable location tracking on your devices? Do you think it makes sense to constantly worry about others learning the truth? I don't.
I want to live in a world where I don't ever have to recall or input a password or a PIN. I want to live in a world where I don't have to lock my doors, lock my bike, lock my PC. I want to live in a world where I don't need to encrypt anything. I want to live in a world where I don't have to worry about privacy settings at all.
I want financial information to be public. I want all media content to be public. I want my location to be tracked and available to anyone at all time. I want my health records to be available to anyone. I want my every thought and emotion to be queryable. I don't ever want to waste energy communicating with the world information that can be tracked by machines.
In the future, people will not fear the NSA. People will not fear online tracking. People will embrace them. We will soon see people pay third-party services to track their every move, their every thought. Why waste my time being explicit about what I want and do when I can delegate that task to something else? The only thing I should input in a system are my values, my opinions, my feelings. Everything else is objective.
Privacy is the antithesis of society. Privacy is an act of pure selfishness. Privacy is slowing down innovation and discoveries in every possible field. Privacy is what inhibits humanity to reach the next level. I can't think of anything that sustains worries and fears more than society's expectation of privacy.
Privacy is a symptom, a defense mechanism. Privacy is not a cure.
I live in an oppressed country. I need privacy to try to topple the government by any means. They watch us, they hear us, they follow us, the read everything we write, they want us dead, we want them dead.
Survival is a harsh word. Learn it or die.
Do you think Tor, Wikileaks, Bitmessage, Darknet are just toys that geeks code in their basements just for fun?
Do you know Snowden, Assange, Stallman, and many others fight for your rights and your liberty against those who try to oppress you?
Innocence, meet reality.
Privacy is more than a human right, is a vital need.
In heaven you don't need privacy, in hell you don't have it. Welcome to reality.
Please understand there's a right time for private secrets to be revealed. Reveal them too soon, when you are not ready yet, and it causes more harm than good.
> I have to wear a mask because everybody else does.
No you don't. It seems like this is the problem you are trying to solve. It's a subtle problem, in that norms sweep you easily, but it's solution isn't shooting down privacy.
Your comment would also be received better, particular the claims to what privacy is, if it contained citations to peer reviewed, published research.
My girlfriend suggests a menstrual calendar app that enables men to predict women's moods, especially the days they're most amenable to amorous solicitation, and generally because it's good for men to know their partners' cycles. A woman might enter the data herself and share it with selected parties, or men could try to maintain their own records on behalf of women who aren't motivated to do it. There is also a monetization potential for advertisers insofar as women are more inclined to pamper themselves at certain times.
Especially since women's vocal range is affected by their cycle. I thought to use DSP to help determine their cycle and facilitate pregnancy or help avoid it. It turns out getting the proper audio baseline for the vocal range of women is tough and they'd be saying weird words into their phone daily.
1. Roll your own Google Maps overlays / plugins for navigation of large facilities. I envision a system where you take a series of GPS readings using your phone around the facility (a few on each floor). You then scan a blueprint or floor map and are able to import the structural data via OCR. Finally you roughly correlate those GPS coordinates with spots on the floor map and the software uses them and the known structural information / layout to create a Google Maps overlay.
2. A peer to peer service for facilitating sharing of paywalled scientific journal articles. You could put in a request for a specific article, the utility would check which of your friends had access to the material, your friend would be prompted to download the article and place it into a designated folder for transmission.
An ifixit[0] like site for repairing your car. I really like the format of ifixit because they give detailed pictures and the tools you will need. A site like this would be great for cars, ranging from just changing oil and tires to possibly harder things.
A Macbook can not crash into another car and kill all the passengers.
I'm not sure that's an actual problem though. The UK had "haynes manuals" for many years. The actual problem is that specialist tools are needed more often now.
Real-time Groupon. All sorts of businesses lose money due to last minute cancellations - give them a platform to advertise discounted slots in the near future.
That's an awesome idea. You can just create an offer in seconds and publish it. And have an app track you based on locations or checkins, remind you once you are in the restaurant.
That's actually been tried by Groupon in the past... It wasn't successful largely because it's hard to get merchants to adopt that way of thinking and also because people don't walk around with the Groupon app open all the time.
Doesn't really need an app, just needs a phone number. Even if your initial marketing list for a business is every customer they've ever had, you're still starting with something.
181 comments
[ 5.0 ms ] story [ 213 ms ] threadLast weeks Idea Sunday: https://news.ycombinator.com/item?id=7616910
2 weeks ago: https://news.ycombinator.com/item?id=7582077
3 weeks ago: https://news.ycombinator.com/item?id=7541601
Humble recommendation: Maybe we can post these in the original post for people who would be interested?
Groups (such as email groups and Facebook groups) and communities need a way to organize and share their data freely, but in a structured way. Also people who have data, or start building some kind of data, in text format, or tabular data, but don't know of any other mean, so they build a blog and start posting everything, then it is impossible to make anything useful with the data later. Or maybe they make a spreadsheet and link to it.
I've come to 2 solutions:
- for the first thing (but also applicable to the second), a structurable database (such as PouchDB) stored in the people's browser and communicating with other people from the group via webRTC could build a decentralized database everyone will have. Some things could be made on top of this;
- other thing is just a blogging platform that is not only a blogging platform (what I really want is it to not be a blogging platform), but it is also some way to publish static atemporal articles, tree-data, tabular data, hierarchical data etc. Written in javascript, running from the browser and communicating directly with GitHub or some other static hosting service with an API. The person would load the contents with the API then deploy the new contents as a static page. Free blogging, no central servers or providers (except for GitHub, but it could be made to generate HTML hostable anywhere), cool stuff can be attached (even some plugin that will take the data from the first solution and share in the blog/page/site).
Pushing and syncing strings is a solved problem, as are atomicity, consistency, isolation, durability. Powerful visualization tools are being made, too. The View and the Controller are taken care of.
The unsolved problem remains: how will we empower users to relate those strings, to have them turn data into information, and information into knowledge? How will we enable the many thousands of non-tech domain experts to create their models? How will we devise the Model?
We wanted to show people that they were in fact able to create structured contents, even if it was in a limited way, and we choose the analogy of file and folders since most people know how to organize them. But structuring and gathering data is half of the problem, you need to translate that data into friendly views and same sort of browsing experience, so we developed same conventions to turn the structure into a small website/embed.
We thought of developing an open way to store and share this data, and let people copy, follow, contribute to these chunks of content. Build a friendly web dashboard that exposes every item with simple views, and which let other developers consume them in any other ways they want.
The problem is that we don’t really know who wants this. We thought that it would be appealing for web agencies as a way to quickly add structured contents to websites, but we were wrong. Some people suggested us to develop a full CMS, but we wanted to work on the problem of sharing and opening up a library of contents, not to be an alternative to Wordpress. We honestly don’t know what to do next. If you want to talk further about it, my email if fsainz/gmail.
- note taking
- server logs
- manual logs of events (I've seen these in use in lots of small business where some employee must log what happened in its turn)
- personal journal
- mixpanel-like thing
- analytics of any kind
- computer-generated logs of anything that doesn't need a full-fledgeded database and system of its own (I'm thinking specifically of the logs of people who used their password to open a dor we have in our building, that communicates via serial port with a server that stores that info, but very ineffienciently)
- (I'll keep thinking about it)
- Online communities. For example, groups of students of some area, groups of pregnant women, groups of people who like tiny houses, groups of people who do geocaching, people who like to save dogs in the streets and give them to adoption, people who like to shoot at birds, people who work with something specific. All these groups probably have groups of email, or Facebook groups, (probably more than one group, maybe various localized groups) where they chat and share interesting links, data and information. Then this data is lost, it isn't searchable by the people who know it exists (specially if they are inside an Excel spreadsheet that was emailed to all the groups' participants), it isn't knowable for people who joined the community later. This people will never know (or will know too later) that that interesting data was there, just behind a pile of emails and posts. There are a lot of examples of data that could be indexed (yes, I'm talking about database indexes) in interesting ways, but I will give none now, and lots of data that could just be organized in an easy way (categories, although I don't like this very much) and made searchable.
For this problem, Trello is a good option, but it is limited, it is centralized, it requires an "owner" of the Trello board to authorize other people, it cannot develop in an organic way, with few people starting to use, then other people joining, and people who don't wanna join never joining. It also cannot index data, just categorize and archive.
- Myself. I don't want a blog, but I have interesting data stored in my computer and my brain I wanna handle to the public over the internet, to whoever be interest in it. I have articles I can't find anywhere else anymore, I have spreadsheet data that should be presented as spreadsheets, I have articles I typed from old magazines, I have videos I don't remember how I got, I also have impressions of books, commentaries about lots of non-mainstream topics, and mainstream topics also, mainly atemporal things, that don't fit well in a blog. I could make a static web page and put all this there, but it seems too difficult for such a low value it will have. I also don't wanna do it because I wanna keep thinking about how would a non-programmer do it. I can't help thinking every people in my situation would: (i) do nothing; or (ii) make a blog, post everything there for two month and abandon the blog. I don't like abandoned blogs.
Pageboxes could do something for this, but I don't know how exactly.
Long time ago, I mused¹ about these kind of things, then worked like mad on devising an implementation. Meanwhile, a side project I’ll take up again, one day.
I love the idea of data blogging.
Thinking more along the lines of what Freebase was doing (before Google smothered it), what Fluidinfo² originally intended to do, and what Silk³ seems to be doing best, right now.
¹ http://rhythmvs.tumblr.com/post/19750973162/hello-world ² http://fluidinfo.com ³ https://www.silk.co
Reading your tumblr post, I think there is useful work to do on the points you mention: Collecting and List Making // Visualising Knowledge // Open Access. We need something simple enough to get the people who already know how to publish posts interested.
Silk is pretty good, and I never heard of Freebase. From wikipedia: "Freebase is a large collaborative knowledge base consisting of metadata composed mainly by its community members. It is an online collection of structured data harvested from many sources". I got to look into this, sounds beautiful.
I love the idea of helping build these community databases and make those contents searchable. I see a challange in the fact that this means adding another channel to discuss contents different than the one they already use, but the promise of making those contents easier to reach might make it worth it.
As for the blog problem, I'm in a similar situation. A blog seems to have some lifespan, and there are some raw contents that I would like to have available without worring about a domain name and the design decisions of CMS theme. I have my own particular "delicious", some comments about different articles, a bunch of photos, and many screenshots and mockups from the different projects I've been working on that I would like just to publish without fuzz (Ideally: add all my project folders. Done). It won't be as beautiful as those web portfolios but I don't need that, I want convenience, I don't like abondend blogs either and I have 3 or 4...
Reading these comments gets me really motivated, thanks!
For a moment, I thought it had been posted as a response for your comment, because it fits well here.
This (http://www.jwz.org/blog/2014/01/psa-back-up-your-shit/) should be good jumping off point to get a lot of the stuff. The system should have customisable hooks written in something like Python to write your own "filters". e.g. "If I've been in emacs for 1 hour straight, pop up a reminder to take an eye break, etc."
Suggestions? Thoughts?
[1] https://www.rescuetime.com/
Now that the smartphone innovation has slowed down a little, why not bring back real buttons, wheels etc.? They should be well-designed, precisely engineered and fun to use. I belive mobile gadget design is now awesome enough to make this work.
I wouldn't mind a scrollwheel, or even real radiobuttons (one button goes in, the other pops up). The exact functionality would still depend on the current application.
Since they give real physical feedback they should objectively be superior to the touchscreen-only functions, no?
http://tactustechnology.com
They are also planning to release it for in-car displays.
This would be great help when trying to dj dancing party: now show me a list of "blues", "chicago", "slow", "with breaks" to choose the next song. Actually in this context, separate service / app would be great - it would work on a mobile device, rather than just desktop.
(a long standing idea on spotify list already: https://community.spotify.com/t5/Spotify-Ideas/Tag-Music/idi...)
Unofficial Google Music API in Python, quite well documented. http://unofficial-google-music-api.readthedocs.org/en/latest...
About to get started writing a little script that will take care of synchronising my library with my google music one, and automatically uploading those that it can't find a match for. A few of my friends also use it, so I was also thinking of making a little tool that would allow us to compare our libraries and show us what each other are missing etc.
Something that could tie in to RightMove (at least in the UK), access your saved properties, and provide a simple interface to schedule viewings (although I guess this would still have to be a manual process to book viewings, but the app should provide a calendar to store them in).
The app could then alert you when the viewing is soon, and provide an interface for you to rate the viewing, add comments and maybe your own photographs to remind you later on, rather than having to scribble notes down on the same piece of paper that becomes home to your chewing gum the next day.
On a social front, it would be good to see how many other people were viewing the same property, to gauge the interest, and maybe you would be able to see an overall rating of a property based on what other people that have viewed think about it?
House/Flat adverts are still relatively useless in the process of finding somewhere suitable, and until you can view and see a property for yourself, it's impossible to tell. I think it would be very useful to be able to see ratings and comments from other users, such as "the second bedroom is a lot smaller than the photos show, and the flooring in the bathroom seriously needs replacing!".
It doesn't look like there's an available API for RightMove at the moment, so I'm not sure how the integration could happen. Still, a stand-a-lone app could still be a worthwhile idea?
EDIT: ...but even then, you might not get personal opinions. Tough one!
I'd personally like to see the use of web/tech to integrate the whole experience - so everything from deposits, rent, grievances and so on is centrally managed and moderated. Again though, hard to get traction.
Social networking is nowadays as central to the internet as email, and it is the first such, heavily used, application layer "protocol" that is closed.
For a small period of time I had hopes that this would be the path G+ would take to the top. It isn't.
I believe integrating social networking into email might be the most promising approach. I've always thought a "ShareBox" icon below my Inbox icon would allow social networking to decentralize in a way that would appeal to end users.
https://tent.io
There are some projects on the tent Github page but nothing with activity less than months old:
https://github.com/tent
I'm not personally involved in the project.
https://wakatime.com
Why? With a taller screen, there's less scrolling and your eyes don't have to jump as much when reading from line to line.
I would like to know if there would be a market for this and why manufacturers phased out 4:3 displays?
There are no "monitors" made anymore just HDTVs. Sometimes real small ones in laptops.
Every time I try to do an in-depth research on Google about some topic I open up 10 or 20 of the most relevant results, only to discover that at least 80% of the information is repeating itself in at least 50-60% of the pages.
One could write a search engine that compares the contents, groups the articles that have similar enough contents and only displays one article per group. This could work especially well in the case of longer searches with enough keywords.
Obviously, this is not a standalone product that would compete against the established search engines, but it could be built on top of them - a browser extension for example.
Any thoughts on this?
The semantic web is the solution. A system that understands content, understands what you know, and understands what you want. Then, all it has to do is show you what's relevant and necessary to the task you're trying to accomplish. In most cases, you won't have to actually do the task, as it will be automated.
I don't see any way around this. Changing how search engines work will still not let you find articles that complement each-other perfectly. And you'll still have to read more than necessary.
>> The service should let you specify topics that you are interested in and will analyze every page you visit (with the exception of blacklisted pages) and put it into the appropriate bucket (topic), for future reference.
Evernote does all of this.
I would pay something like $100/year for such software that would keep up with every new version of its supported programs.
Wine is fantastic but too limited at the moment.
(and since you're on the web, might as well save notebooks, share, edit concurrently in realtime, etc)
One system: Take a photo of the bike every time you park it in a non-standard place (standard being e.g. "work", "home" or "girlfriend's"). Unfortunately, taking out my smart-phone every time I go grocery-shopping is bothersome, and I might only find out I need to suddenly go elsewhere when I've already parked the bike.
Ideally, my bike (or, through statistical inference, my smartphone) should just publish GPS coordinates. Unfortunately this drains the battery very fast. Perhaps a background service that uses the GPS a little more sparingly than the Google Maps app?
Problem: I've got 2TB of videos, music, and images on my hard drives, and I've found my available tools to be too low-leverage on managing this anymore.
Context: about 2 years ago, I've stopped the directory structure madness, and embraced search as first-order link for media files. As a byproduct of this, I notice I don't have a clear mental structure of what videos/music is downloaded to my computer anymore, and to where (good riddance!). Neither do I want it to; instead, I'd like to have a "youtube for local hard drive":
Basically, I'm looking for a local Media Browser application for windows: specifically, an app which allows:
* thumbnailed preview of all videos & images
* Automagically categorized based on all online, or within-file available meta-information
* With a built-in search
To perform the following operations:
* Preview a thumbnail on video files (similar to file explorer)
* Display categorized lists of all media available locally on my drives
* Channelsurf, similar to youtube: be able to start from one movie file, and have "similar" videos around it
* Double-clicking on the file to start mplayer
* typing the first 2-4 letters of the movie/image to instantly bring up the list of media with that part in it
While specifically not doing:
* Anything that requires me to touch any of the files in any way shape or form whatsoever
* Any upsale (I'm looking at you, windows 8 "personal videos"), advertising, or introduce any friction during interaction
* Bringing up any browsers: solution must be windows-native, and responsive (<20ms results)
Specific things I've tried so far:
* The closest thing I can wave towards "want" is apple's itunes, and JRiver's media center; they both share the same weakness of not being able to use mplayer as playback engine
* Banshee on windows is unable to show video previews, have multiple stability issues, and can't use mplayer for double-click
(I've posted full details of this to http://softwarerecs.stackexchange.com/questions/2646/media-b... a month ago, to no response avail, but getting upvotes, which evidences a market to be had for this )
Regarding built-in media players, not one vendor takes usability of playback seriously. Specifically, when I hit "enter" on a file, that means I'd like to see it played back, like, now. Not "queueing", nor 5 seconds later. When I scroll left & right, that means I'd like it to go +-1 min / hour, like, now; and not 5-10 seconds "seeking". On my quadcore 3.4ghz battlestation, this remains a problem for every single native, and web-based app. MPlayer is, to date, the only one which consistently delivers on sub-second videoplayback, and it does so from cold start. (There are also major problems with how subtitles are handled by players, but I grant you that might be a niche)
Hence my strong requirement above for the app to go with shell.exec, instead of half-baked non-implementation of every single video codec under the sun.
edit: Also, I'm not sure how to make it a viable business.
I want this on a set top box... And anything like yamj or xbmc absolutely chokes on my collection and I spend hours fixing incorrectly identified files.
agree with the idea/problem
(I tried FastMail, Atmail, Rackspace, and none of them is a good alternative).
Why not?
- no nice mobile app
- does not integrate with third-party systems (yea "not their problem" but they should do something to promote 3rd party system to integrate with them)
- spam filter is not so great (false positive is the problem - I'm ok having occasional spam in inbox, but email from wife linking some stuff I need order is really annoying.)
It is kinda hard to pinpoint the main reason...
[1] http://perpetuall.net/
I'm aware writelatex and sharelatex exist. I would like a latex feature in Docs though. Preferably one that compiled/rendered the output in realtime
A completely private social platform with messaging, email, timeline, blog, pics, docs, etc.
A completely transparent social platform with messaging, activity tracking, content management, knowledge storage, wishlist/preference/opinion management, idea generation, etc.
You might not know yet, but privacy is the root of all evil, and the privacy madness has to stop.
The more we hide, the more we have to hide. It's a never ending cycle.
People don't see the world as it really is. They see the world the way others want it to be seen. They see an artificial, curated version of the world where people have no flaws.
This disconnection between what is real and what we think is real makes people feel broken, different, not good enough. How are they supposed to know that nobody is perfect when everybody actively try to show their perfect side? We all wear masks by fear of rejection, fear of being different.
Just like you, I need privacy because privacy is expected of me. I need to lie because people don't want to hear the truth. I have to wear a mask because everybody else does. Being completely transparent and honest is a sure way to be rejected and jailed. But if I had the choice, I would prefer to be myself, to never have to remember lies, to never have to hide.
Do you think the privacy paradigm is sustainable? Do you think it makes sense to use so much resources to encrypt everything? Do you think it makes sense to hide yourself behind proxies and firewalls? Do you think it makes sense to protect everything with passwords? Do you think it makes sense to keep creating fake profiles online? Do you think it makes sense to disable location tracking on your devices? Do you think it makes sense to constantly worry about others learning the truth? I don't.
I want to live in a world where I don't ever have to recall or input a password or a PIN. I want to live in a world where I don't have to lock my doors, lock my bike, lock my PC. I want to live in a world where I don't need to encrypt anything. I want to live in a world where I don't have to worry about privacy settings at all.
I want financial information to be public. I want all media content to be public. I want my location to be tracked and available to anyone at all time. I want my health records to be available to anyone. I want my every thought and emotion to be queryable. I don't ever want to waste energy communicating with the world information that can be tracked by machines.
In the future, people will not fear the NSA. People will not fear online tracking. People will embrace them. We will soon see people pay third-party services to track their every move, their every thought. Why waste my time being explicit about what I want and do when I can delegate that task to something else? The only thing I should input in a system are my values, my opinions, my feelings. Everything else is objective.
Privacy is the antithesis of society. Privacy is an act of pure selfishness. Privacy is slowing down innovation and discoveries in every possible field. Privacy is what inhibits humanity to reach the next level. I can't think of anything that sustains worries and fears more than society's expectation of privacy.
Privacy is a symptom, a defense mechanism. Privacy is not a cure.
Prove me wrong.
I live in an oppressed country. I need privacy to try to topple the government by any means. They watch us, they hear us, they follow us, the read everything we write, they want us dead, we want them dead.
Survival is a harsh word. Learn it or die.
Do you think Tor, Wikileaks, Bitmessage, Darknet are just toys that geeks code in their basements just for fun?
Do you know Snowden, Assange, Stallman, and many others fight for your rights and your liberty against those who try to oppress you?
Innocence, meet reality.
Privacy is more than a human right, is a vital need.
In heaven you don't need privacy, in hell you don't have it. Welcome to reality.
> I have to wear a mask because everybody else does.
No you don't. It seems like this is the problem you are trying to solve. It's a subtle problem, in that norms sweep you easily, but it's solution isn't shooting down privacy.
Your comment would also be received better, particular the claims to what privacy is, if it contained citations to peer reviewed, published research.
Especially since women's vocal range is affected by their cycle. I thought to use DSP to help determine their cycle and facilitate pregnancy or help avoid it. It turns out getting the proper audio baseline for the vocal range of women is tough and they'd be saying weird words into their phone daily.
2. A peer to peer service for facilitating sharing of paywalled scientific journal articles. You could put in a request for a specific article, the utility would check which of your friends had access to the material, your friend would be prompted to download the article and place it into a designated folder for transmission.
[0]: http://www.ifixit.com/
EDIT: If you create the site, please let me know. This is not something I can work on right now but if it's a wiki like site I'd love to contribute.
I'm not sure that's an actual problem though. The UK had "haynes manuals" for many years. The actual problem is that specialist tools are needed more often now.
http://www.ifixit.com/Device/Car_and_Truck