| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | ||
| 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| 13 | 14 | 15 | 16 | 17 | 18 | 19 |
| 20 | 21 | 22 | 23 | 24 | 25 | 26 |
| 27 | 28 | 29 | 30 |
GSA makes the case for open source
Coleman said that while GSA has hired a company to support the instances of Linux in its data center, the agency saved money by discontinuing support of its knowledge management software, KnowledgeTree, because the software was stable and the agency no longer required a lot of changes. By doing so, the agency was able to drop the $20,000 per processor support fees.
Navy to focus only on open systems
By Peter Buxbaum, Published on March 6, 2008
The Navy will acquire only systems based on open technologies and standards.
Vice Adm. Mark Edwards, deputy chief of naval operations for communications, broke the news March 5 to a Navy IT Day audience in Vienna, Va., sponsored by AFCEA International.
“The days of proprietary technology must come to an end,” he said. “We will no longer accept systems that couple hardware, software and data.”
The Navy’s decision was informed by a combination of motivations, including the desire to provide the latest capabilities to warfighters and control the costs of its information technology operations, he added.
“We can’t accept the increasing costs of maintaining our present-day capabilities,” Edwards said. “In the civilian marketplace, it’s just the opposite. Some private-sector concerns are cutting their costs by 90 percent while expanding their performance.”
Edwards noted that the Navy has cut the number of databases and applications it maintains and has reduced its networks by 40 percent. “But it is not enough,” he added. “We would have to double our IT budget over the next several years just to run in place.”
By using an open network architecture, the Navy could rapidly upgrade its capabilities and handle increases for demand, Edwards said.
“Above all, we must break the stovepipes of data so that we can share information across domains,” he said.
The failure to lead in technology could have dire consequences, Edwards said.
“The situation is very similar to that of the Soviet Union at the end of the Cold War,” he said. “Because we put them in a position of always having to catch up, the mere threat of the Strategic Defense Initiative crippled the Soviet ability to continue the arms race and enabled our side to dictate terms. If we remain behind in technology, a future adversary will eventually bring terms to us.”
Disaster Tech: What is working and what is coming
Jesse Robbins (O'Reilly Radar), Mikel Maron (Mapufacture)
Twitter and Google Maps are being used in mainstream emergency management, and projects like InSTEDD will push them even farther. This session shows you what is working, what isn’t, and what’s next in Disaster Tech.
...
pattern: disaster tech innovation
1. disaster
2. ad-hoc adaptation
3. championship
4. iterative improvement
<breathe>
- Katrina
jesse g robbins
worked with world shelters
was in hancock county Mississippi
all street signs were down, etc.
ad-hock adaptation: relied on google maps - bad because the main bridge was out
Jesse Robbins:
Problem: Google Maps isn't realtime
bridge out, then not
works with openstreet map (OSM) - uses RESTful interfaces
working with UNJLC
UN considering using OSM
lots of stuff done on the fly in disaster, tech needs to be able to rapidly adapt
San Diego fires:
- 900k evacuees
- 500k acres burned
- 1500 homes destroyed
phone went down, but SMS worked
Nate Ritter's Twitters, twitter.com/ritter
hashtags.org
ARC now using twitter twittger.com/redcross
Champion: InSTEDD - SMS ReoChat
twitter is a tool of first resort because it it there
InSTEDD, working with traditional Disaster Agencies to build new tools
Tsunami:
225k dead
Ad-Hoc Adapation:
JRC tsunami model
1. usgs georss earthquake feed
2. eu commission lightweight tsunami propagation model
3. georss polygons republished
Champion:
GDACS.org: global disaster alert & coordination
they went with GeoRSS as a solution
Iteration:
Pending.
- will it work and apply to publishing alerts
savings lives means losing lives
shouldn't be scared away by the fact you can't save everyone, must take risks
ANTI-Pattern:
the search for Jim Gray
Champion: none.
public now believes that this is easily repeatable
next iteration:
Steve Fossett search
- inadequate trainging
- many false positives
- people called SAR teams directly which hindered search (- no feedback to people when they submitted a 'hit')
- didn't have new and old imagery to compare
Maj Cynthia Ryan (air patrol) to paraphrase - this didn't work
need to actively learn lessons and apply them and create champions
New group stood up:
www.internetsar.org - helping with search and rescue on the internet
mikel@mapufacture.com - mike maron
jesse@oreilly.com - jesse robins
Joel Selanikio (DataDyne.org), MD pediatrics
Just saw a really great talk by Joel Selanikio from www.DataDyne.org at Etech2008 about the need to focus ICT development in Africa on mobile devices first versus browser based desktop.
Also check out this mobile partnership between MIT and Nairobi University: http://eprom.mit.edu/ - EPROM, part of the Program for Developmental Entrepreneurship within the MIT Design Laboratory, aims to foster mobile phone-related research and entrepreneurship.
.....
The Case for Africa as a Mobile Development Hothouse, Joel Selanikio (DataDyne.org)
http://en.oreilly.com/et2008/public/schedule/detail/1310
Whats happening in africa with cell phones and the interent
why different
need to increase capacilty in Africa (ablity to develop, etc.)
1994-2004, 58% subscriber growth - itu 2006 report (americas 23%)
reasons: starting at a low point
cell phones becoming ubiquitoius, UN and everyone else missed it, the market responded and created made this hapen
they are using unlocked iPhones as well
Cell phone and internet expansion
2004 - 82 millions people using cell phones
2007 - 4.7 internet users
- completely different way of communications occurring
- need to use the infrastructure they have in place
- people assume they eventually will go desktop browser base
- 2 fiber lines down the coast of Africa's + satellite is the access to Internet
- Africa needs more fiber
Access will continue to be different for US and Africa, Cell phone versus US-Internet
- low bandwidth cell phones is the most likely route to the internet for the foreseeable future
what can you do with slow processors and low-bandwidth?
Nathan, MIT has a collaborative program with Nairobi University, they end up working to troubleshoot Windows/Vista, etc. Need to get them to develop technology for their countries versus picking up after Windows
For Example:
How to 'text internet' via SMS, give special groups (doctors, etc.) to get drug dosage, etc.
Managing your medical record via SIM cards (shot records, etc.).
Send reminders to parents to get vaccinated
text messages are cheap, if you can drop say polio huge savings and life saved can occur.
help them develop their own applications
Application in Senegal fishermen buy insurance via SMS before they go out
Mo Ibrahem, Sudanese, runs one of the biggest cell network in Africa
Mobile banking . M-Pesa in Kenya. Cell phones in Africa are mostly prepaid, this has become a market by using cell phone minutes for money. Enables people without access to a bank to access what a bank gives you, the ability to move small amounts of money around.
South Africa called mizit (?)
International development want to make things better in developing countries but don't want to make money, They also focus too much on the desktop browser experience versus other means of comm's
DataDyne (non-profit) started to move data collection to PDA's, uses mobile programmers in Nairobi versus US.
Working with the Vodaphone Group Foundation and the UN Foundation to setup/support programs to increase development capacity for mobile phones
- http://eprom.mit.edu/ as a model and potential partner (Nairobi)
- setting up a dev center in africa to support public health programming needs
Need to open up the market
Joel Selanikio
joel@datadyne.org
The Case for Africa as a Mobile Development Hothouse
Joel Selanikio (DataDyne.org)
8:45am - 9:30am Thursday, 03/06/2008
General Marina Ballroom F
Africa is leading the world in year-over-year growth in mobile penetration, and other parts of the developing world are close behind. Most of the people who are now gaining access to cell communications and Internet via cell phones have no other method to access the Internet: their paradigm for Internet use involves mobile devices and small screens, unlike developed country users that generally use cell phones for voice calling, with Internet access being an occasional activity.
This dependence on mobile devices for Internet access means that developers in Africa and elsewhere in the developing world can see the mobile device with fresh eyes: not as a replacement for a desktop, or an imitation of a laptop, but as a platform in its own right. There is a very high chance that they will “see” possibilities that we, with our long history of computing, are blinded to.
By helping to foster a community of developers in developing countries, I believe we can reap tremendous dividends, with new perspectives being applied to old problems, and with the potential for as-yet-unimagined applications.
Disaster Tech: What is working and what is coming
Jesse Robbins (O'Reilly Radar), Mikel Maron (Mapufacture)
2:50pm - 3:35pm Wednesday, 03/05/2008
Twitter and Google Maps are being used in mainstream emergency management, and projects like InSTEDD will push them even farther. This session shows you what is working, what isn’t, and what’s next in Disaster Tech.
...
Disaster Tech: What is working and what is coming
Jesse Robbins (O'Reilly Radar), Mikel Maron (Mapufacture)
2:50pm - 3:35pm Wednesday, 03/05/2008
Mission Hills
Twitter and Google Maps are being used in mainstream emergency management, and projects like InSTEDD will push them even farther. This session shows you what is working, what isn’t, and what’s next in Disaster Tech.
...
pattern: disaster tech innovation
1. disaster
2. ad-hoc adaptation
3. championship
4. iterative improvement
<breathe>
- Katrina
jesse g robbins
worked with world shelters
was in hancock county Mississippi
all street signs were down, etc.
ad-hock adaptation: relied on google maps - bad because the main bridge was out
Jesse Robbins:
Problem: Google Maps isn't realtime
bridge out, then not
works with openstreet map (OSM) - uses RESTful interfaces
working with UNJLC
UN considering using OSM
lots of stuff done on the fly in disaster, tech needs to be able to rapidly adapt
San Diego fires:
- 900k evacuees
- 500k acres burned
- 1500 homes destroyed
phone went down, but SMS worked
Nate Ritter's Twitters, twitter.com/ritter
hashtags.org
ARC now using twitter twittger.com/redcross
Champion: InSTEDD - SMS ReoChat
twitter is a tool of first resort because it it there
InSTEDD, working with traditional Disaster Agencies to build new tools
Tsunami:
225k dead
Ad-Hoc Adapation:
JRC tsunami model
1. usgs georss earthquake feed
2. eu commission lightweight tsunami propagation model
3. georss polygons republished
Champion:
GDACS.org: global disaster alert & coordination
they went with GeoRSS as a solution
Iteration:
Pending.
- will it work and apply to publishing alerts
savings lives means losing lives
shouldn't be scared away by the fact you can't save everyone, must take risks
ANTI-Pattern:
the search for Jim Gray
Champion: none.
public now believes that this is easily repeatable
next iteration:
Steve Fossett search
- inadequate trainging
- many false positives
- people called SAR teams directly which hindered search (- no feedback to people when they submitted a 'hit')
- didn't have new and old imagery to compare
Maj Cynthia Ryan (air patrol) to paraphrase - this didn't work
need to actively learn lessons and apply them and create champions
New group stood up:
www.internetsar.org - helping with search and rescue on the internet
mikel@mapufacture.com - mike maron
jesse@oreilly.com - jesse robins
in case you missed this WSJ article. Really interesting about a guy trying to build a shoe business. the problem is that he can't buy small quantitates of finished materials. It really reminds me of the reverse side of what JSB talks about, if you have a good infrastructure network business thrives, but the feedback loop goes the other way as well.
I am sure that the defense industry is having this problem as well... and will get worse
....
U.S. Shoe Factory Finds Supplies Are Achilles' Heel
By TIMOTHY AEPPEL
POMPANO BEACH, Fla. -- Howard Shaffer opened a small factory here in 1995 to make sneakers for Adidas AG -- but also to make a point.
Having spent the previous decade setting up plants in China to manufacture shoes for big U.S. brands, he thought he knew how to revive the moribund U.S. footwear industry: use heavy automation run by a handful of skilled workers instead of relying on large numbers of low-paid Chinese laborers.
..
What killed his U.S. factory isn't just competition from Asia's cheap labor, he says. It is the lack of infrastructure needed to make a factory tick, a problem that has bedeviled the few remaining independent shoemakers in the U.S. Finding technicians to fly in on short notice to fix shoe machines was a constant and growing challenge, Mr. Shaffer says, because the number of U.S. companies that make and service machines has dwindled. The suppliers of shoelaces, leather and other basic materials insisted that he buy in batches far larger than made sense for a small-scale producer.
Steve Cousins
- Willow Garage out of Stanford
- Private Robot Ver 1 (PR-1), can go around any ADA compliant bldg. so far only tele-operated
- all open source software
- have produced 10 PR2 robots as a platform for researchers
- people need the ability to tinker
this continues to prove that (eventually) good (and useful) technology eventually becomes open source
Chris Anderson
www.diydrones.com
- split up navigation and stabilization (buy stabilization for $ 80, uses infrared - looks for horizon gradient)
- connected to LEGO mindstorms for stabilization (~$1000)
- use cell phone for navigation and video (~$500)
would be good if DoD could go back to building systems like this
Peter Norvig:
- algorithms for 'squishing' of pictures (from sigraph)
- internal coherence for images and people tends to work better by clustering results
- text problems: mistakes were made in translation, 98% accurate for words
- spelling too
- Google Sets: use word sets to find other sets
overall pretty interesting view on how to make the best prediction with massive amounts of data, with various tricks. Moving into translation as well
Programs such as the Army’s Future Combat Systems and organizations such as the Defense Information Systems Agency and the Defense Intelligence Agency are using open technologies using non-proprietary software.
A roadmap for the adoption of open technologies was released last year by the deputy under secretary of defense for advanced systems and concepts. That paper proposed adopting open source infrastructure and technologies and applying open source to collaborative technologies being implemented by DoD.
One of the advocates of the open source approach is Brigadier General Nickolas Justice, Army program executive officer for command, control and communications tactical.
Justice acknowledges that many in the military acquisition community remain wary of open source. "They are concerned about the security of open source products as well as the level of maintenance they require," he said.
