[Fiware-i2nd] Collecting feedback to Reviewer's report on I2ND GEis

[Banniza, Thomas-Rolf (Thomas-Rolf)]

Hi Pier,



As Altoclient was not mentioned specifically in the Schubert report, please find below only some answers/comments to the general recommendations:



recommendation #1  is to ensure that the GEIs are easy to find and use. This means not only that the documentation needs to be complete and fully comprehensive, but also that the implementation is automated to a maximum degree - ideally supported through automatic installers. To ensure integration into existing frameworks, they must however be configurable to a wide range of existing setups. In this context, the documentation must make absolutely clear what is supported and how. The instructions should involve clear steps for the setup and configuration - some GEIs already provide such steps, yet to a different degree of completeness. Different failure causes must be identified and clearly elaborated. Ideally, working demo installations are made available with a clear tutorial on how to use them and how to configure the own infrastructure. Keep in mind that the end user is not familiar with the product and potentially not even with the according dependencies.

Summary Recommendation #1: Easy and configurable installation packages with comprehensive documentation



                Comment: In my eyes, Altoclient documentation is clear and describes how to install Altoclient, this includes also a script which can run (as far as useful) automatically.



recommendation #2 to update the catalogue to maintain a clear and comprehensive list of all the GEs and their implementations. The current catalogue is not only incomplete, but makes it very difficult to clearly associate GEs with their implementations. It must be easy to find implementations on basis of their functionality, and not on basis of their name. In this context, it would be good if clear reference use cases / demonstrations are provided, to ensure that the user knows not only how to use the software (recommendation #1) but also to what purpose. The catalogue must thus act primarily over the GEs and for each one list all reference implementations, along with their respective status. If a specific implementation is discontinued, clearly state so.

Summary Recommendation #2: Elaborate the catalogue to allow identification and retrieval of GEIs on basis of their functionality



                Comment: To allow a purposeful search in the catalogue, the catalogue as a tool lacks a more elaborated search function(and also fields/tags in the GEi templates which can be filled with purposeful text or something similar. This is a genuine task of the catalogue tool provider and catalogue manager. What concerns more specifically the Altoclient GEi, the core functionality (retrieval of network information) can be  easily found when playing with the existing catalogue search function. Basically, this holds also for the other NetIC GEis.



recommendation #3 to clarify the "impact" from software usage for each GEI. This means in particular that the requirements, the impact from installation on the infrastructure and the legal setup of the dependencies is made clear. Keep in mind that in a commercial setting, all concerns legalistic and economic must be absolutely clear - in other words: the licenses must be defined, the hardware and software requirements must be clarified (ideally for different usage scenarios, considering scope), the conflicts with typical other setups must be elaborated etc.

Summary Recommendation #3: Clarify the consequences from installation in maximum detail, involving in particular requirements and legal implications from dependencies (licenses etc.)



                Comment: In my eyes, the installation and administration guide contains all technical  information needed to install the Altoclient GEi. What concerns the 'economic and legalistic issues': on the one hand, the terms and conditions are predefined by FIWARE to at least 90%, so any critics on their usefulness should be sent to TID (or who did this?). On the other hand: it is 'worldly innocent', if a reviewer expects from research partners 'absolutely clear' economic information. A research result is not a product. And before a company offers a product, it has to undergo a lot of checks inside the company (because this accroaches a lot of, e.g., responsibility and reliability issues) , and this would probably not be done in the frame of a research project.





Best regards,

Thomas

































Thomas R. Banniza

ALCATEL-LUCENT DEUTSCHLAND AG

RESEARCH ENGINEER

BELL LABS, GREEN TOUCH PROGRAM



Lorenzstrasse 10

70435 Stuttgart

Telefon: (+49) 711 821 32232

Fax: (+49) 711 821 40147

Thomas-Rolf.Banniza at alcatel-lucent.com



Alcatel-Lucent Deutschland AG

Sitz der Gesellschaft: Stuttgart · Amtsgericht Stuttgart HRB 4026 Vorsitzender

des Aufsichtsrates: Michael Oppenhoff

Vorstand: Wilhelm Dresselhaus (Vorsitzender) · Hans-Jörg Daub · Andreas Gehe



Diese E-Mail-Nachricht und ihre etwaigen Anlagen können vertrauliche

Informationen enthalten. Wenn Sie diese E-Mail-Nachricht irrtümlich erhalten

haben, bitten wir Sie, uns dies mitzuteilen und die E-Mail und ihre etwaigen

Anlagen sowie etwa hiervon gefertigte Kopien umgehend zu vernichten. Die

irrtümlich erhaltene E-Mail und ihre etwaigen Anlagen dürfen weder

weitergeleitet noch anderweitig verwendet werden.



This e-mail and its attachments, if any, may contain confidential information.

If you have received this e-mail in error, please notify us and delete or

destroy the e-mail and its attachments, if any, immediately. If you have

received this e-mail in error, you must not forward or make use of the e-mail

and its attachments, if any.















-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.fiware.org/private/old-fiware-i2nd/attachments/20131212/01c5dc97/attachment.html>