Experiences in Crowd Sourcing
ARwJgR Of course, I understand a little about this post but will try cope with it!!...
Strange but true. Your resource is expensive. At least it could be sold for good money on its auction!...
Thank you, a very interesting note!!...
TwitHawk
Elisabeth Oppenheimer
(Formerly TwitterHawk, but Twitter made them change the name.)
The TwitHawk space looked forlorn, so I took a quick look at it. It's pretty clever. You set up a search -- say, people tweeting about Amazon Mechanical Turk within 25 miles of Palo Alto -- and the app updates you when someone tweets on that topic. You also set up some prepared responses ("Iâm studying Amazon Mechanical Turk for a class, how has your experience with it been?") which are automatically sent to anyone who matches your search. You can also change the settings such that you need to confirm before sending your message, or so that you automatically follow anyone you message. It's basically a marketing tool: in the example they give, a coffee shop searches for local people tweeting about a desire for coffee, and then sends them the address of the shop.
The concern in the press coverage is that this will massively spam Twitter users, but the owners have taken some useful steps to prevent this--you have to pay 5 cents per message, you can only send one message to a given user, and you can only send one message every two hours.
I'm impressed with the service, although Iâm still fundamentally stunned that Twitter and Twitter apps have morphed so quickly from a "huh? why would you want to do that?" phenomenon to a "cornerstone of all cool marketing campaigns" phenomenon.
Reuben Rodriguez
First things first. As mentioned above, Twitterhawk is dead. Long live Twithawk! I think Elisabeth did a pretty good job of describing what Twithawk does so I'll keep my description to the bare minimum: Twithawk is basically a tool that sends preset replies to other Twitter users whose tweets match whatever search results you have set up. Twithawk gives new registrants 10 free auto-replies, so I set up a new Twitter account and gave it a spin.
I decided I would be tweeting as HarryDoyle97 and accordingly constructed a search for any mention of the word "Indians" within 25 miles of the Cleveland area. I created three pre-set responses and set up the search to run once every two hours. Within 10 minutes, Twithawk had sent out my first auto-response (JUST a bit outside) to Indians on Fan Feedr in Cleveland, Ohio. It was all very easy.
Elisabeth discusses the critics' main concern about how Twithawk could be used for mass Twitter spam and she mentions the steps Twithawk has implemented to try and combat this possibility - the five cent per reply cost and the limited frequency with which a response can be sent. There is also subtle encouragement to use the Twitter feed as more than just a robot automatically replying to search results. Each "campaign" that you set up has a meter on your dashboard labeled "TwitterHawk Noise Ratio" with an arrow pointing from green to red. It then rates the ratio between Twithawk responses on your feed vs. "natural" (read: by an actual human being) responses and encourages the user to keep the ration below 25% (1 Twithawk response for every 4 real ones). While this may not deter those with the real intent to spam, at the least it serves as a helpful reminder to the well-meaning, but inattentive marketer to keep their feeds from being mere spam. Finally, the FAQs asks those who feel that Twithawk is being used for spam to report the spammers and notes that Twitter is active in punishing spammers. If Twithawk remains vigilant, the product definitely has potential.
Tyler Lacey
I used TwitHark too. My goal was to find people complaining or wondering about on the status of the (often late) Boston subway and bus system called "The T" and run by the MBTA. So my search was simple: I merely searched for terms such as "MBTA", "T", "late", "delayed", etc. that posted within 5 miles of Boston. My response was to tell people about the official MBTA twitter feed (MBTANow) that already supplies people with very up to date information. I set TwitHawk to send this response automatically.
I suspect that my response will annoy more people (at most 10 really, since I am only using my ten free responses) than it helps for two reasons. The first is that "within 5 miles of Boston" is too coarse of a search area. It will likely define lots of people that do not or can not use the T and exclude many people that do use the T. This difficulty is a symptom of the my second, more general, complaint with TwitHawk's search mechanism: it is too simplistic. For something as delicate as searching the millions of twitter posts and responsing in a useful manner, I think I need more than the simple list of terms that are presumably searched for in an "or" fashion, narrowed by a single location. I would prefer something more akin to regular expressions (http://en.wikipedia.org/wiki/Regular_expression) that would let me specify precisely the strings that I am searching for, as well as a more precise manner of specifying locations (so I could use something like "within 1 mile [Town]" for every town near Boston that the T serves). This way I would be able to target posts that are more likely complaning or inquiring about the status of the T and search within an area that more closely resembles the T network rather than a circle extending out from the center of Boston. I believe that this advanced search capacity should be built into TwitHawk using a more sophisticated search builder interface. This would allow users not familiar with the complexities of regular expressions to make use of their flexibility. However, an API or program that sat on top of TwitHawk could achieve a similar result by flattening a regular expression into many simpler searches and automatically entering those searches in TwitHawk. However, for complicated regular expression searches this would create an unmanageable number of simpler searches for TwitHawk to store.
I also noticed that a captcha is the only thing that seemed to be standing between me and multiplying my ten free opportunities to spam into millions of free spam messages by automatically creating TwitHawk accounts and setting them to the same searches. Without knowing the internals of TwitHawk I can't be sure what defenses exist to prevent this.
Drumbeat Privacy Project
Michael Feldman
Ok. So I was intimidated by some of the other possible projects such as Twithawk and Mechanical Turk. Signing up for things and following directions? Ew. So, the clear solution was to try and tackle the creative problem proposed by Drumbeat to simplify various internet companies' privacy policies, a la Creative Commons. The idea is to create a simple structure of icons, images and phrases to more effectively communicate a companies privacy policy without requiring the user to sift through a ton of dense legalese.
In creative commons, each icon which stands for a particular set of rights the author wishes to protect. But this won't work for privacy policies because the policies are different for each company. They don't merely invoke a set of already-delineated rights. One organization called Privacy Choice has already tried to create icons to simplify the process of learning about companies privacy policies, (see http://www.privacychoice.org/whos_watching). But this website is not completely effective since it requires the user to "mouse over" various categories of privacy policies, only to encounter much more "mumbo jumbo."
It seemed to me that the primary design problem was that there are multiple levels of information: specific data within specific provisions within general categories. So I tried to create an idea for a basic system which used the Privacy Choice Categories (plus a few others I thought would be relevant), color coded them so that a user could tell which categories might require greater attention, and then placed the provisions on a rather vague scale. While not as specific as an ultimate solution might want to be, this at least gives a user site-specific information that will be useful before delving into the actual text. From here, the next step would be to create a stock of short phrases which could appear when the user mouses over the icons which would give more detailed information without resorting to large blocks of text.
All in all, the challenge is very interesting and definitely something I would be happy to continue to work on over the next few weeks. The specific challenge of conveying all information in a given privacy policy in simpler symbols is one which will likely require involvement of the actual companies to conform their policies to set rubrics (or at least stock modular blocks of text), but short of that, it might be possible to create a more intuitive set of symbols that could be more generally applicable to all policies. But then there would still be the problem of aggregation of information since people would have to go through each policy to decode it and re-encode it into the new system.
You can find the actual suggestions I posted to drumbeat here: https://wiki.mozilla.org/Talk:Drumbeat/Challenges/Privacy_Icons.
Bruno Magrani
If Firefox is about making the web better through software, drumbeat intends to be the twin brother that focus on connecting project ideas and people together to promote a better web. In a way, one can think about it as an incubator for projects to improve people's experience on the web.
Because the project seemed quite exciting to me I took the liberty to exchange some emails with the people involved to get a sense of how I could be of most help to the project and decided to collaborate coming up with other examples of people who are contributing to improve the web. This first initial experience â being able to chose how to best participate in the project â denoted one of the wonders of online peer production: the advantage over traditional industrial information production systems in terms of information gains regarding how to connect a person's interest with a specific task.
If one important component of the project involves coming up with ideas for projects an equally important component involves getting real life examples of who is doing what to improve the web. It works both as way to recognize the efforts of these people, but also to stimulate other people to do the same and start their own initiatives. In a sense this is intended to send a message saying: everyone can and should contribute to make a better web.
One of the things that I notice while reading the projects and people being featured was that they were too much focused on experiences happening in the United States and some in Europe, so I decided to come up with examples of what both common and notorious people were doing in a developing country such as Brazil. My contribution can be found here and I'm definitely going to keep contributing to the project.