Monday, June 30, 2008

The Battle of the Forges, Part II: 'And the winner is ...'

Last Friday 27th I attended an OSOR meeting at Brussels, and things have cleared a lot.

OSOR's goal is to become the central observatory and repository of both code and information from and for EU public administrations. It is to be born with the aim of freedom of use, so that most of its tools will be simply of volunteer use. It thus clearly becomes the solution for gisEIEL.

Nevertheless, as Mancomún is devoted to promoting the use of FOSS in Galicia, which is gisEIEL home country, we feel that we should keep our project registered with them. This will introduce a little 'noise' in our managing tools. But anyway! Will try to cope with it!

So, our solution is using just gisEIEL project own collaborative tools, despite what's the language used (be it Galician, Spanish or English), thus building a multilingual environment that we believe that will perfectly fit the actual reality of our target users/collaborators. As an example: at the above mentioned OSOR meeting, at least five different languages were used (English, Spanish, Italian, French and Danish -and maybe also Irish-), despite the fact that English was the official one. And everybody seemed to feel quite comfortable with that situation, moving on-the-fly from a language to another, with total naturality.

So, summarizing: our next steps will hence be:
  • Registering at OSOR
  • Negotiating the (un-)use of Mancomún tools
  • Going on incorporating new members to ourcommunity (by the way: more news will follow on this issue soon)
And the winner is ... (drums sounding): "There's no winner, her Majesty!" (An America's Cup classic :-D)

Monday, June 16, 2008

The Battle of the Forges, or where should I post my project?

If you remember past entries, we have registered gisEIEL project at Mancomún and Sourceforge forges, as a way to facilitate the access to it from users coming from different regions (Galicia and the rest of the world, in this case) But now we face a problem: each forge does automatically launch its own series of supporting and collaborative tools. Result: our project "enjoys" now three bugtrackers, three sets of mailing lists, three code repositories, three svn and two sets of fora (project's site lacks of this tool).

And the question is that those tools cannot be linked one to each other. I mean, our will would be that we could set a link from Mancomún and Sourceforge bugtrackers to our own. And the same for mailing lists. But this seems not to be possible.

But the problem doesn't end here. Now, a new forge has come to town: OSOR, the European Commission's Open Source Observatory and Repository. It deals with the task of fostering the use of OS software by European administration bodies and, as a consequence, its responsibles want administration born projects to be served from their forge. And gisEIEL is administration born. And it is addressed to administration bodies use. And, of course, it has been invited to register at OSOR.

But, once again, OSOR will provide the project with new sets of collaborative tools. So, we face now the need to take the decission of registering or not at OSOR. And to do so, we will have to ballance the convenience of supporting OSOR initiative (High) against the need for new, quadruplicated collaborative tools (None, indeed)

So, the decission is ... (drums sounding ... .... .... ...

Ehmmm! Well! Keep in contact and you'll know!

Tuesday, June 10, 2008

Diputación de Pontevedra adopts gisEIEL

Diputación de Pontevedra (Pontevedra Province Council) has also adopted gisEIEL as their GIS tool to manage EIEL geodatabase. In some weeks, we will hold a meeting to transmit to them our know-how on that matter and to coordinate our databases, given that both provinces share borders one with each other

Friday, June 6, 2008

About project's size

Wondered about what's our project size? Do you want to know it prior to downloading code? Ok, no problem: here you have some figures about it.

Data generated using David A. Wheeler's SLOCCOUNT on gisEIEL and webEIEL

(SLOCCount, Copyright (C) 2001-2004 David A. Wheeler - SLOCCount is Open Source Software/Free Software, licensed under the GNU GPL)

1.- gisEIEL:

  • SLOC Directory SLOC-by-Language (Sorted)
    • CatalogoMetadatos
      • java=38517
      • jsp=1174
    • EIEL-FichaMunicipal java=32931
    • EIEL-Formularios-Alfanumericos java=17672
    • EIEL-GeneracionFormularios java=10251
    • EIEL-GisEIEL java=8714
    • EIEL-Validaciones java=3547
    • EIEL-Informes java=1344
    • RemisionDatosEIEL java=982
    • EIEL-GestionDeMapasGisEIEL java=948
    • EIEL-GestionPermisosGrupos java=570
    • EIEL-Autenticacion java=526
    • EIEL-Utilidades java=482
    • EIEL-DeGeomASimpleGeom java=443
  • Totals grouped by language (dominant language first):
    • java: 116927 (99.01%)
    • jsp: 1174 (0.99%)
  • Total Physical Source Lines of Code (SLOC) = 118,101
  • Development Effort Estimate, Person-Years (Person-Months) = 29.98 (359.81)
    • (Basic COCOMO model, Person-Months = 2.4 * (KSLOC**1.05))
  • Schedule Estimate, Years (Months) = 1.95 (23.40)
    • (Basic COCOMO model, Months = 2.5 * (person-months**0.38))
  • Estimated Average Number of Developers (Effort/Schedule) = 15.38
  • Total Estimated Cost to Develop = $ 4,050,493
    • (average salary = $56,286/year, overhead = 2.40)

2.- webEIEL

  • SLOC Directory SLOC-by-Language (Sorted):
    • EIEL-Web
      • java=7736
      • jsp=7374
      • php=114
    • EIEL-VisualizadorSVG java=8688
    • EIEL-Nomenclator
      • java=4952
      • jsp=91
    • EIEL-ServicioNomenclator java=2794
    • EIEL-AWMS java=1541
  • Totals grouped by language (dominant language first):
    • java: 25711 (77.23%)
    • jsp: 7465 (22.42%)
    • php: 114 (0.34%)
  • Total Physical Source Lines of Code (SLOC) = 33,290
  • Development Effort Estimate, Person-Years (Person-Months) = 7.93 (95.20)
    • (Basic COCOMO model, Person-Months = 2.4 * (KSLOC**1.05))
  • Schedule Estimate, Years (Months) = 1.18 (14.12)
    • (Basic COCOMO model, Months = 2.5 * (person-months**0.38))
  • Estimated Average Number of Developers (Effort/Schedule) = 6.74
  • Total Estimated Cost to Develop = $ 1,071,695
    • (average salary = $56,286/year, overhead = 2.40).

Hope this helped! :)