Problem with Brand New SwiftRiver Install (URGENT — FOR PHILIPPINES DISASTER RELIEF!)

[David Schlesinger]

Hi, I got involved in Geeklist's Typhoon Haiyan Hackathon this past Monday morning as a result of reading the coverage in TechCrunch.

Since Monday morning, I floated an idea to put up an Ushahidi, which turned into a project, which resulted in a site which can (we think) address a request from the President's Communications, Development and Strategy Office. I delivered documentation on this last night to them, and am waiting to hear feedback.

In order to do a better job of managing the Twitter stream for the Ushahidi site, I installed a SwiftRiver v1.0 using the appropriate branch on github and following the documentation on the Ushahidi wiki. Everything's gone fine, modulo a couple of minor tweaks, but I'm unable to log into the site using the credentials "admin" and "password". I've tried a variety of things, up to and including poking different values into the "users" table in the database, but I can't get it to work.

Can anyone help? I notice that the login dialog requests an email address rather than a user name, but the default value for that doesn't work, nor do different values that I plugged in by hand from the MySQL command line.

This is completely blocking my ability to upgrade the data feeds on the Ushahidi. I've got an email out to Emmanuel, but haven't seen a reply so far. Any help or suggestions would be very much appreciated.

[Chris G. Blow]

David, 

I suggest you do not commit to any urgent work with this codebase at this time.

AFAIK Swift is officially unsupported now — the wiki instructions must be out of date.

I think this list would surely be interested to know more about your interests, and there may be others who want to collaborate with you on your specific design problem.

Chris Blow

--
You received this message because you are subscribed to the Google Groups "SwiftRiver" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swiftriver+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[David Schlesinger]

I'm not planning on committing any work, urgent or otherwise, to the code base at all. My goal was to provide something that could potentially help out relief efforts in the Philippines as part of the #hack4good "hackathon" being sponsored by Geeklist.

However, it's become increasingly apparent that this effort is largely a waste of everyone's time, there's no sign at all that anyone associated with Geeklist is actually in meaningful contact with any agency or organization in the Philippines in a position to take advantage of any of this.

My goals — since stock Ushahidi only provides the ability to pull in tweets based on hashtags — were to use SwiftRiver provide a more flexible for Ushahidi to mine the Twitter stream.

My immediate goal is to be able to log into the silly damned thing, so I can see how it works, but as I've indicated, that seems impossible, in spite of multiple efforts on my part to force it to let me in.

I've found archived forum messages indicating an identical problem from about a year and a half ago, with no response or resolution. I've attempted to contact Emmanuel and/or anyone at Ushahidi.com in a variety of ways, here, on Twitter, on LinkedIn, via direct email, without response.

It would be great if organizations might put something on their repositories advising people that the code is "unsupported" — not to mention no longer listing it as though it were an active product on their web sites. Evidently, I've been wasting my time and spinning my wheels on this.

Know of a supported open source project that does what SwiftRiver was meant to?

[Hapee de Groot]

Hi Chris,

I know since Heather went to OKFN this group has more or less been silent but the news that Swift is AFAYK officially unsupported now is kind of shocking to me, since when? and how officially is it? Can anybody else comment on this?

Kind regards,
Hapee

----- Original Message -----
> David,
>
> I suggest you do not commit to any urgent work with this codebase at
> this time.
>
>
> AFAIK Swift is officially unsupported now — the wiki instructions must
> be out of date.
>
>
>
> I think this list would surely be interested to know more about your
> interests, and there may be others who want to collaborate with you on
> your specific design problem.
>
>
> Chris Blow
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> On Nov 14, 2013, at 10:55 AM, David Schlesinger <
> stone...@gmail.com > wrote:
>
>
>
> Hi, I got involved in Geeklist's Typhoon Haiyan Hackathon this past
> Monday morning as a result of reading the coverage in TechCrunch.
>
>
> Since Monday morning, I floated an idea to put up an Ushahidi, which

> turned into a project , which resulted in a site which can (we think)

[Giuseppe Calamita]

Hello, I'm available to install Sfwitriver for a deployment related to th Philippines Typhoon so to check if the same issue will happen to me and then trying to fix it; I'm looking for a free web host service to install it and I wonder if you may suggest me one tha may features: Python, Tomcat etc. all of the packages required for installing Swiftriver (given thw wiki is updated). In this video tutorial I noticed there should be a installer too: https://www.youtube.com/watch?v=up0Y-VemWZA

Regards,

Giuseppe CALAMITA
twitter: @cypherinfo
www.skillsforchange.com

[Giuseppe Calamita]

Sorry the coreect link is this: https://vimeo.com/12579692

Giuseppe CALAMITA
twitter: @cypherinfo
www.skillsforchange.com

On Thursday, November 14, 2013 9:55:33 PM UTC+1, David Schlesinger wrote:

[Stone Mirror]

Thanks for your help, folks, and I'll look into that tutorial.

I'm going to continue to play with SwiftRiver out of personal interest, but I've gotten a separate message from someone from Ushahidi, indicating that because of the $500 cost of "the Twitter firehose", the project as a standalone thing was essentially abandoned as being too costly for anyone to use.

I have to admit, I'm not really sure I understand this: I'd imagined that this wasn't vastly different than using the Twitter API, and a synthetic Twitter ID perhaps, to follow certain other IDs and do searches. I don't really know much about the Twitter API at the moment, other than having set up the conduit "app" for Ushahidi. I suppose I'll have to learn a little more. Any enlightenment on this would be appreciated...

(This should be a simple problem, though, right? Unless authentication is Just Plain Broken in v1.0, or I've bobbled something, then I should be able to plug in a row in the users database with the correct values and get in, right? So what's the 512-bit MD5 hash for "password" or some other Well Known Value...? Hm. An Approach, perhaps.)

Anyway, it's something of a moot point now, since this "hackathon" has now ended, victory has been declared. Unfrotunately, I haven't seen any actual evidence of any of this hitting the ground or assisting a single person or organization actually working in the Philippines right now.

I sent out detailed documentation on the Ushahidi as it stood very late last Tuesday (or early Wednesday) in response to this "request from the President's" etc., etc., and have never received any response, not so much as an acknowledgement.

While I was waiting the organizers pressed me to do what amount to busy work — e.g. putting up a "microsite" dedicated to solving the same problem I'd already solved with the Ushahidi, despite the total lack of evidence that anyone in the Philippines had actually looked at the site.

One of the program managers also shared with me a story about someone from Doctors Without Frontiers "finding" the site — a site which had no domain name associated with it at the time, just running off a naked IP address — and being "disappointed" to find no medical information on there. When I pointed out the complete unlikelihood of this, especially considering that my actual email address is right on the home page of the site and no one from DWF had gotten in touch with _me_, he admitted that he'd "misspoken".

I've come to the conclusion that this was essentially an ill-thought-out publicity stunt — it did, in fact, get them two quite breathless articles in TechCrunch — no one involved seemed to have either meaningful contacts in the Philippines nor any previous experience with a disaster relief and recovery situation.

Many of the projects were extremely unrealistic — projects predicated on the possession of a smart phone and a working Internet connection when the Philippines has the lowest penetration of smart phone ownership in the entire Asian region, for example.

A learning experience to be sure. Sites like Tomnod did a great job of harnessing the "power of the crowd" to do stuff like identify different kinds of damage on satellite imagery; others classified tweets. That worked _great_. Harnessing the power of a crowd of developers? Not so much.

Seems like crowd-sourcing is a lot better for tightly constrained gruntwork than actual development and deployment of complex software in a complex situation. Live 'n' learn.

Thanks for the help!

--
You received this message because you are subscribed to a topic in the Google Groups "SwiftRiver" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/swiftriver/ZD386Il8G8U/unsubscribe.
To unsubscribe from this group and all its topics, send an email to swiftriver+...@googlegroups.com.