Preparation DL2014 and review demos wiki page

Sunday evening

There are quite a few people arriving before the Monday so we've arranged a meeting spot for anyone who wants to get together on the Sunday evening. We'll be meeting at the Holborn Whippet near Kings Cross from 6 PM that evening. They have a bar menu and serve food until just after 9 PM so people arriving a little later can still get something to eat. The map location is here, the nearest tube station is Holborn about 150m away or it's a 30 minute walk from Kings Cross if the weather's nice.

The workshop

Venue: The Great Hall, College Building, City University London's Northampton Square campus - http://www.city.ac.uk/visit#9541=1

First part of the workshop will be a panel session with Ross, Ed and David Giaretta (and Hildelies as moderator). After that a quick 1-minute-only Elevator pitch session for all projects before the 'Clinic' part with the stalls.

There will be at least 13 project stalls of which SCAPE has three. Normal stalls will probably be 2X2 M, but we hope to have one a bit larger for our Table 3 (Tools).

A draft of suggestion (from our side) of how the room will look like has been sent to the organizers Room_Setup.jpg^ ^ More info about rooms etc. can be found in the info mail that was send to the other projects involved in the workshop here.

If you want to know more about the planning and which other projects are invited, check SCAPE-APARSEN Workshop September 2014 and subpages Workshop description + Proposal review

Clinic theme

The workshop has a sustainability and a clinic theme. There will be a special clinic table where people from different DP projects can give advice and answer questions. We will have a schedule with 2 or 3 people per slot of 15 minutes for this. We will ask SCAPErs for this table. Please let Jette know if you can man this table for a slot!

SCAPE Stalls

There will be 3 “tables” of demos. These 3 sets of demos will also form the basis for demos at the EC review.

We want to rent screens to show the SCAPE demos and other visual presentations, like these: https://www.flickr.com/photos/dhbenelux/14549975735/ and  https://www.flickr.com/photos/dhbenelux/14549975805/ which can be used instead of a table.

Please add all posters and leaflets to this folder on the Sharepoint before the 15th of August the latest, but please do so earlier if possible!  

https://portal.ait.ac.at/sites/Scape/Shared%20Documents/Forms/AllItems.aspx?RootFolder=/sites/Scape/Shared%20Documents/Presentations/DL2014%20Workshop%20posters%20and%20leaflets

Table 1: Integration

Overview: PW lifecycle demo around an image migration scenario with execution on Taverna Server

Other SCAPErs needed at this stall: NB: Do they know what is expected of them?

Melanie

Other things needed for this table: please fill additional things needed!

Laptops, one specially powerful to be able to run a VM with the demo installed.

Table 2: ‘Scalability’ focussed

Overview:
This is a TB focussed table, providing scenarios of how scalable execution could be employed and showing what sort of results can be/have been achieved (compared to non-scalable execution), rather than specifically demonstrating  hadoop running.

Other SCAPErs needed at this stall: NB: Do they know what is expected of them?

*Other things needed for this table: *please fill additional things needed!

Table 3: ‘Tools’ and Sustainability focussed

Mainly based around the online demonstrators (http://scape.opf-labs.org/)

Other SCAPErs needed at this stall: NB: Do they know what is expected of them?

Ed

*Other things needed for this table: *please fill additional things needed!

Old info:

Posters:A leaflet and poster template is available here. The leaflets should follow the template. The posters should follow the template as closely as possible -at least everything under the line should be according to the template layout. 
The logos in .eps can be found here and here 

Overall strategy for the posters

What happens very often is that people see a poster and think: oh, this is not for me. Then they walk to the next stall/poster. To avoid this we need to make our audience understand what we are about in seconds.
This applies to all posters, also to the so called technical ones.

NB: The SCAPE intro poster is very high level, don't expect that one to solve all problems for the non-technical audience for you :-).

Some points:

• Your poster needs to explain in non-technical, short, catchy, easy to read language what is in the poster/ a particular part of the poster, like: ‘know what you have’ or ‘save time’ as headers.  
• Please use as less words as possible and keep the poster as empty as possible 
• To get to the core of what you want to tell through the poster, you can try to put this on paper in no more than 3 sentences (not bullet points). This is often rather difficult, but this is what the poster should bring across.
• Please use Melanie as try out audience: if she doesn't understand what you want to say we have to adjust parts of the poster. This can often be achieved by a few simple tricks like different headers, with which TU can help.  

Table 1: ‘Integration’ focussed: PW lifecycle demo around an image migration scenario with execution on Taverna Server

Initial idea presenters: Michael, Luis, Melanie 

Demo preparation team:

• Kresimir
• Luís
• Melanie
• ?

Link to flow chart end to end demo 

Dissemination material:

• Demo
• Poster
• Screencast

Identified task list demo 1  (directly from the post its developers workshop - not edited yet):

• Define use case
• Portable Demo (VM)
• Update poster
• Screencast
• Collect reasons for this scenario
• Make sure we have a small dataset for the demo that people are allowed to take home
• C3PO support for Data Connector API
• Define control policies -> easy to evaluate /using tools
• Readable text as notifications (email)
• Find better term for Planner
• Data Controller API to allow sending of ? events with Q&A output
• Finish integration of Taverna & include Q&A info
• View plan in RODA
• Plan with better graphics
• C3PO/Scout should only see the normalized files
• Output: Visual! Plus workflow (code Taverna)
  

Table 2: ‘Scalability’ focussed: Poster based

Initial idea presenters: Sven, ?, ?

Demo preparation team:

• Peter
• Sven/Martin Schaller
• Rune
• Will

Overview:
This is a TB focussed table, providing scenarios of how scalable execution could be employed and showing what sort of results can be/have been achieved (compared to non-scalable execution), rather than specifically demonstrating  hadoop running.

Posters:
The idea would be to have posters to cover the following:

• (Possibly four) Poster(s) describing a scenario from each TB (or one big poster)
• Poster about how scalability is achieved
• Poster(s) showing results of scalable execution (with comparisons against non-scalable approach)

Demos:

• Raspberry Pi cluster running hadoop to attract people to the stand and as a discussion point around hadoop (Carl is working on getting the Pi’s working).
• A FITS/ToMaR/C3PO web archiving demo on a laptop (screencast as backup)

Points to consider (directly from the post its developers workshop - not edited yet):

• Explanation explaining linkage between Table 1 demo (integration) and these scalable demos?
• Ensure consistency in scenarios between demos:
  • i.e., use the same scenario (TIFF->JP2 migration) as for integration demo
• Results should highligh appropriate benefits of using scalable approach, e.g.:
  • Compare scalable and non-scalable approaches
  • Faster than before
  • What is now feasible that wasn’t before
  • Commodity hardware instead of expensive
• How scalability is achieved could/should be presented in a non-technical way
  • e.g. counting coins example? Other?
• Showing hadoop output isn't that exciting
  • Can it be visualised better, e.g. load on each computation node?
  • Use Pi's as attractor to stand, and as something to discuss hadoop/scalability around
• Perhaps give ‘Under the hood impression’, for instance ‘webcam’ images of data centres

Dissemination material:

• Posters
• Raspberry Pi cluster
• Laptop with Web Archiving demo
• Screencast (as web archiving backup)

Identified task list demo 2 

• Scenario Posters:
  • Decide on whether to have Research/Data Centre TB scenarios
  • Create LSDR Poster
  • Create Web Archiving Poster
  • Identify Research/Data Centre TB scenario
• Scalability Posters
  • Create poster about how scalability is achieved
• Results Posters
  • Collate scalability results for LSDR and Web Archiving scenarios
  • Create poster(s) with results
• Web Archiving Demo:
  • Setup standalone laptop running FITS with ToMaR & C3PO hadoop demo
  • Ensure backup screencast is also viewable (just in case)
• Raspberry Pi Demo:
  • Decide on whether to use Pi's or not, if so...
  • Setup Pi cluster running Hadoop
  • Create an appropriate, long running hadoop job
  • Test robustness of cluster
  • Create Plan B: screencast of Hadoop output

Table 3: ‘Tools’ focussed

Mainly based around the online demonstrators (http://scape.opf-labs.org/)

Initial idea presenters: 

Demo preparation team:

• Carl
  See up to date table above.

List of tools is based on the available online demonstrators, for now:

• Pagelyzer
• Matchbox
• Jpylyzer
• Xcorrsound
• Flint
• DICOM Tools
• hawarp
  

Dissemination material:

• Online demo per tool
• Poster per tool

Identified task list demo table 3 

• Poster per tool - to be made by the main ‘owner’ institutes (based on the leaflets?)