HKML stuff

some older correspondence put here for reference:

Candidasa Prabhu,

Please accept my humble obeisances.
All glories to Srila Prabhupada.

> —–Original Message—–
> From: On Behalf Of Candidasa dasa
> Sent: Sunday, 4 September 2005 8:07 AM
> To: ‘Discussion about ezine preaching’
> Subject: [Ezine] high quality + detailed summaries
>
> Hare Krishna,
> > Information overload.
> I’ve been thinking about this issue.

Nice to hear. Me too. I’ve been thinking about it since 1999 and am happy that I have some time to finally do something.

> Yes, having billions of classes on
> available on the Internet will not solve anything. It will just bewilder
> everyone by giving them too much choice. However, if these billions of
> classes are as well organized as Amazon’s product selection, with detailed> class summaries, transcriptions, devotee reviews, related blog postings,> then more is better.
>

I agree and disagree with this. I don’t see any problem with having lots of classes available via a low-level organizational scheme such as by speaker (the old HKAL way). I certainly think that we should embrace higher levels of organization though as you suggest for sure but these can be added as we go. My main immediate concern is that something is put up that can serve as the core for later expansion.

Separating content from presentation lends itself quite nicely to modular expansion. I think the ITC folks are going this direction with Drupal integration which I’m happy with. I don’t want to become dependent on these guys for everything though as I think that its unlikely that our needs will forever run parallel with theirs. I especially don’t want to rely on them for server space and plan to dedicate a server to HKML probably around mid next year or maybe sooner depending on donations I receive for it.

> Lots of choice isn’t a bad thing, we all have slightly different needs,
> desires, likes and dislikes. It’s just too much choice without a means of> making an intelligent choice that bewilders people: 150 different brands> of
> toothpaste, a gazillion different TV channels, etc. However, no one
> complains about there being too many webpages on the Internet. After all,> we> have Google.
>

Yes. So I was very interested in some kind of data-matching ontology thingy as you were introducing to us early on. I hope that something can still be done like this even if its not as full-blown as we were hoping.

> So then: new (potential) prime requirements for the HKML:
>

Good. Now we are defining our standards before development. (Much more intelligent than after the fact.)

I think we are due for some definitions in terms of the different roles we are going to play in HKML development and implementation. Also who is onboard here? In what capacity can they contribute? If we are going to recruit developers from the devotee community then we have to have some mutually agreed upon roles defined first. I’m not confident that we have come to any mutually agreed upon conception of the project yet – what to speak of defining roles to the point of getting others involved. I’d rather have something functional albeit basic up and working and have a ratified conception of the HKML established first so that we can systematically define roles and tasks. Then we won’t frustrate any potential contributors by being disorganized and changing our game plan on them after they’ve expressed interest.

Since we’ve begun these dialogues, the HKML the project has metamorphosized quite a lot. I attribute this to having several devotees interested in this and putting our heads together. I hope that the definition and development phases can reap these cooperative benefits. For the sake of clarity I’ll be overt in my intentions to head up this project. Its been my meditation for over 6 years now and I’ve invested a lot of time and energy thus far. I have also organized my family, career etc… to allow for me to focus on this until some time next year so that’s what I intend to do. If anyone has an issue with this I’m happy to discuss with them and intend to listen to the experience and know-how of others which often far surpasses mine.

> – High quality recordings! If something isn’t high quality, it doesn’t get
> included. Devotees need to have confidence that everything in the library
> sounds decent. No one will miss a few low quality lectures.
>

True.

> However, even a high quality recording still needs some post-processing.
> The> audio editing process:
> Beginnings and endings trimmed, volume-leveled, peaks normalized,
> jaya-radha-madhava cut out, verse-chanting cut out, end-of-purport-reading
> and beginning-of-questions bookmarked, noise reduced (gated), encoded at
> 64> kbps and ID3-tag added.
>

Good. Only thing is I think that 32kbps/mono is a better target as it’s a more manageable file size and plenty good enough for classes. We’d need to make no-brainer tutorials available for everyone.

Maybe you didn’t get the email but I wrote an outline earlier of how we could simultaneously keep HKML pukka with a reliable selection of classes as well as accommodate the deluge of classes that are surely not going to conform to our standards. This is the direction I plan to take so if you see any problems with it than feel free to point it out.

Here’s an illustration of the model I have in mind:

The default presentation of the site is that just the core classes are presented. Users can include the outer ring of uncertified classes via a filter if they choose to do so. These classes will then show up in searches/queries but will be clearly demarked as UNCERTIFIED in the interface. Users can still include these in their customized playlists/podcasts.

> – Detailed description/summary! A class labeled “SB 8.23.05: Material
> world”
> is not going to spark my interest to listen to it. Each media item needs
> to
> have a detailed description of what is being said and who the audience is
> (this should be at least one or two paragraphs of text).

I think we should accommodate this sort of description but I don’t expect that everyone will spend the time to describe each and every class like this. Some will be happy to take this up as a form of meditation but not everyone will have the time. For instance, I think it’s a great idea and would like to do that for each and every one of the 1000’s of classes I have on my hard drive but I don’t even have the time to listen to all of these classes what to speak of write a 2 paragraph summary.

We should accommodate that sort of thing though for those who are inclined. If we have a community of contributors then this kind of work can be farmed out to whoever wants to do it. As you said – the more metadata the better for each file. These descriptions can be added after the file is available though, eh?

I also really was interested in that technology you pointed out to us that automatically transcribed the audio file and made it searchable. Ie – http://www.podscope.com/search.php?q=blog

So the general idea I’m headed towards is a centralized core of organization and standards while accommodating other conventions and standards in a less focused way.

>Not only does
> this> help people find items they want to listen to. It also helps the person
> writing the description remember the Krishna conscious nectar they have
> just> listened to.
>
> Do you agree/disagree?
>

Agree.

Your servant,
Ekendra das
http://www.gopala.org/
—————————
shriman-mauktikadama-baddha-chikuram susmera-chandrananam
shri-khandaguru-charu-chitra-vasanam srag-divya-bhashanchitam
nrityavesha-rasanumoda-madhuram kandarpa-veshojjvalam
caitanyam kanaka-dyutim nija-janaih samsevyamanam bhaje

I worship Sri Caitanya Mahaprabhu, who is being served by all His devotees and associates; whose hair is bound with strings of pearls; on whose moonlike face is the nectar of His gentle smile. His beautiful golden body is covered with lovely garments and various shining ornaments. He is so charming, being absorbed as He is in the enjoyment of sweet mellows in dancing, and is more splendid in His dress than even Cupid himself.


Posted

in

by

Tags:

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *