Development of Librarian Software?

  • Hey gang, I just wanted to follow up on a comment, Kempermaniac posted in another thread. Since I don't want to further derail the thread, I decided to start a new one. :)


    What do we know about the problems of developing e.g. a librarian? Nothing do we know about it, that's the simple truth. And if somebody of us knows, why don't you create it?


    You're right, we don't know. But my take on this is that maybe we should know. :)
    In other environments, metadata is well documented. Think about ID3 tags in MP3 audio files. Or think about EXIF or even XMP metadata in image files. It's relatively easy for those with some programming experience to come up with software to edit, analyze, handle these kinds of metadata. If there's a clear distinction between the metadata block and the actual profile/rig data, there's no need to publish the structure of the core data. Just the metadata structure could be documented and published and thus made accessible to all kinds of developers.


    Of course I'm not in the position to know or judge on Kemper's internal business plans. If they decide to keep exclusive control over every aspect of their product, that's probably fine .... although I tend to believe this would be a mistake regarding the metadata. There are so many smart people here that could potentially extend the Kemper universe with some amazing tools, from web applications to smartphone apps to computer software that facilitates rig management. Often these kinds of 3rd party tools even provide the respective mothership with great ideas, for free.


    We've seen CK's comment on swarm intelligence ... I won't totally nail him on the wall for this. But still I think he underestimates the potential creative power of a highly convinced and motivated swarm.


    One last comment on this topic for now:
    I think it would be pretty easy to reverse engineer the metadata structure and come up with some tools. Afair, there had been one project already. But I won't jump on this boat unless Kemper decides to officially open up this specific path.


    Hope I could at least correct some of your comment, Kempermaniac. :)


    Cheers,
    Martin

  • Hope I could at least correct some of your comment, Kempermaniac.

    Hey, everything's fine. :)
    My point wasn't the librarian software. It was just an example and must be seen in the whole context of my post. Personally I would love to have a librarian software or/and an editor.
    And as a side note: It's not that easy to make every point as clear as I want, because I'm no native speaker.

    I could have farted and it would have sounded good! (Brian Johnson)

  • I have created a lot of toolkits and even firmware mods in the past.
    So its not a problem to create an editor or librarien.


    Its just a lot of work - and I have lost interest after the announcement from CK that the mothership develops their own one soon.

    (All trademarks are trademarks of their respective owners, which are in no way associated or affiliated with soundside.de)


    Great Profiles --> soundside.de

  • HI,


    A long time ago, i started to write a tag editor , some notes can be found here : https://dl.dropbox.com/u/74887749/kpa_tag_inspector.html
    it was simply providing a way to edit tags from rig files, waiting for a more complete product from the mothership, but as I got almost no feedback about building a platform independent installation package, i stopped its development.
    Like Armin said, it's a lot of work


    More information can be found in an old thread called "Basic Rig Tags Extractor/Editor" in the "Introduce Yourself" part of the public forum.



    Laurent :)

    Laurent


  • So, Laurent, now might be a good time to say thanks! I've been using your tag editor since i got my Kemper and it's a lifesaver. :thumbup:
    Seeing as you got the tag reading and saving done, it was a pity development stopped. It worked brilliantly and the only thing missing from it was XML import/export and we could do batch-editing - then we wouldn't have had to wait for Kemper to release their own solution.

    "But dignity is difficult to maintain
    stamina requires constant upkeep
    repetition is boring
    and you pay for grace."

  • It worked brilliantly and the only thing missing from it was XML import/export and we could do batch-editing - then we wouldn't have had to wait for Kemper to release their own solution.


    You can export to a csv-format. And you can do batch edit when you edit the tags in that csv using Excel or something the like. I'm away from my studio computer so I don't know the actual commands but i did use that quite a lot for editing amp names to fit in my sorting scheme. I would copy and paste or autofill all the tags in Excel and then save the csv file and do the batch processing. I once batch processed over 1.500 files.

  • Totally appreciate the work Guy did on that app.


    I think that if Batch is too difficult, maybe adding Drag n Drop would be a huge time saver. Or remembering what folder you were on.


    But could you guys explain the cvs and excel bit more? I'd like to give that a go.

  • Another +1 of gratitude for the work that was done on the editor. I used it a bunch to edit tag names of my favorites once I'll culled the list to sub 300 profiles. Simple stuff like changing some of the wacky Amp Factory bundles amp naming to make sense and be more easily sorted. Why Andy didn't name the amps used by Make/Model/Year/Mod still baffles me.. For example I had to added "Marsh" in front of the amp name "JCM 800 Hot Mod" so when i'm sorting by amp name, all the Marshalls are together instead scattered all over the list. Did this for approx 80 profiles from various makers. With out the editor I would have been forced to use the Kemper encoder knob method, one... letter.. at... a... time.. Basically, I would have just said screw it, and not bothered if that was the case.


    A batch editor if possible would be golden, but the basic editor did the job for me..


    Just wanted to say thankyou to the original makers! :thumbup:

  • Totally appreciate the work Guy did on that app.


    I think that if Batch is too difficult, maybe adding Drag n Drop would be a huge time saver. Or remembering what folder you were on.


    But could you guys explain the cvs and excel bit more? I'd like to give that a go.



    The principle was to :


    1 - Give the tool a directory containing rigs to be processed so that it parses all rigs and generates a csv file containing all tags of all rigs
    2 - User imports this csv file into excel and performs all desired changed (rename , copy , paste tags thanks to excel power)
    3 - Give the tool this new excel sheet as an input : every rig found in this excel sheet is copied and its tags are changed using the ones found in the excel sheet.


    Don't know if my explanations are very clear, not very fluent in english...
    Please let me know,
    Laurent

    Laurent

  • The reason the tag editor didn't get tons of feedback is it's a very limited tool as compared to a librarian. editing tags is really only helpful to those who are creating lot's of profiles. For the players consuming profiles rather than making them, the tag editor isn't a tool many would use.


    A librarian would be nice, but then again what exactly does one mean when we request it? editing a backup file can be done today rather easily. What librarian features are we hoping for? A more robust ability to edit backup files or something more in which the KPA is connected to a computer? This would require 2 way communication with the KPA which might take some time to think over and decide exactly how to implement it


    For an editor, this would also require 2 way communication with the KPA with potentially even more integration. Someone might be able to accomplish most of this today with some type of NPRN editor. Again, while there are people who likely sit a KPA on a desk and keep it constantly hooked to a computer there are other users like myself who would be happy never to connect it to a computer.


    This is one of those hot button topics which could potentially result in a lot of anger and frustration from folks whose high expectations are not met...

  • Ah, that makes sense. I thought that maybe there was a way to do that now. My confusion.


    Thanks for the explanation. Your English is perfectly fine!


  • I don't know. I do regular gigs and also session work -
    so long as performance mode doesn't work, i might need 5 or 10 variations of each amp with names that are appropriate to song parts for live use, and also the stock amp, dry, under it's own name for use in the studio.


    That means all the profiles i got from the rig exchange have to be tagged consistently and that i'd want to load up different profiles based on what gig i'm doing - all are easily doable with Laurent's tool and simply impractical with the KPA's interface.

    "But dignity is difficult to maintain
    stamina requires constant upkeep
    repetition is boring
    and you pay for grace."

  • The reason the tag editor didn't get tons of feedback is it's a very limited tool as compared to a librarian. editing tags is really only helpful to those who are creating lot's of profiles. For the players consuming profiles rather than making them, the tag editor isn't a tool many would use.


    Will, with all respect, will you please consider not everybody is using his/her Kemper in a way you would imagine?


    Laurents tool is incredibly useful for people who USE a lot of profiles. Since the filter and sort options of the KPAs OS appear and it was possible to sort by amps, a consistent tagging was a bare necessity. Amp tags are all over the map, even for the factory profiles (they never listed the real amp for a start).


    So basically those sort options are plain useless if you don't have your tags sorted in the first place. Now do that with 1.000 profiles and the KPA interface...

  • Some more food for thought regarding the librarian software development/design. :)


    1. What do we need now, what could we need in the future?
    That's an important question when a company has to plan new features or general enhancements. When it comes to the core of librarian software, first task is to define all possible requirements for now AND the future, because future redesign of software is way more difficult than good initial planning. So let's jump in for a moment. What could be important for a KPA user? For example which guitars a specific rig is good for. One rig can be perfectly fine for more than one guitar. So it's not enough to have one guitar tag field with one possible value only. Multiple values would be very helpful, which immediately makes sorting, categorizing, searching inside the Kemper a more complex task for the OS. Also, if Kemper realizes how important metadata is to organize content, then it's a pretty obvious task to make this metadata model as flexible as possible for future extension. You simply can't afford to create a completely new metadata model every year, right? Flexibility is key and this needs thorough planning ahead of implementation.


    2. What do we need on the Kemper, what on our computer/laptop/device?
    That's also an important question to answer. There might be lots of possible information that only matters in a librarian software while it's not needed on the KPA. Examples? How about images and/or audio examples available for each amp rig in your librarian software? How about lots of different tags (even custom tags) to categorize your rigs by gain, age, preferred music style, commercial vs. free profile, etc.
    I think you'll notice immediately that not all of this information makes much sense on the KPA itself while it would be a GREAT help inside your librarian software. So there's potentially some reason to split rig files into those used on the KPA and those stored in the librarian's rig database. Of course you could always export to the KPA required format, stripping all the unnecessary data.


    3. Edit your rigs until you drop dead!
    I mean, even if you buy your rigs from Andy or Armin, you will most likely edit them to your taste / requirements. And it's good and important that you actually can. But still it might be problematic if you need to keep track of the rig's origin in some way. So it would be nice to keep the original rig data inside the librarian, like in a second (original) metadata layer, while you edit and tweak as hell in your user metadata layer. The original metadata layer wouldn't even be important on your KPA if you already entered your own metadata. But you would still be able to always get information about the original creator.


    4. How does a user find the rigs he's looking for?
    Wow, that's something very important as well. Basically that's why we're asking for a librarian. So we need all the means to filter our rig archive. "I have a rig archive of 4,000 rigs but now I need a Fender with a tube screamer and delay for my ES-335." ... You don't want to do this search by file name, you need custom search filters. And wouldn't it be great if you could save these search filters for quick re-use? Many things possible, even if it's to much for a first version of a librarian software. But the foundation for these things has to be there, so it can be implemented whenever it seems necessary.


    I know this is to much thought for some who just want to use their KPA and play guitar. But that doesn't mean that these things shouldn't be discussed and considered (at least internally at Kemper). And these thoughts are one of the reasons why I won't try to create a small tag editor like the one that has been created before. It just lacks the flexibility and the multimedial features to really nicely handle (find, rate, select, sort) your rigs on a computer.


    Cheers,
    Martin

  • I don't know. I do regular gigs and also session work -
    so long as performance mode doesn't work, i might need 5 or 10 variations of each amp with names that are appropriate to song parts for live use, and also the stock amp, dry, under it's own name for use in the studio.


    That means all the profiles i got from the rig exchange have to be tagged consistently and that i'd want to load up different profiles based on what gig i'm doing - all are easily doable with Laurent's tool and simply impractical with the KPA's interface.


    I do the same thing. Store the amps with different names in the device. Are you actually changing the tags or just the profile name?


    Will, with all respect, will you please consider not everybody is using his/her Kemper in a way you would imagine?


    Laurents tool is incredibly useful for people who USE a lot of profiles. Since the filter and sort options of the KPAs OS appear and it was possible to sort by amps, a consistent tagging was a bare necessity. Amp tags are all over the map, even for the factory profiles (they never listed the real amp for a start).


    So basically those sort options are plain useless if you don't have your tags sorted in the first place. Now do that with 1.000 profiles and the KPA interface...


    Of course everyone uses their KPA differently, I would never assume differently. From the author's mouth, he got no feedback. Chances are not many folks were using it. Are you actually changing the tags or just the name? I totally agree the renaming via the interface is a bit tedious, but do you really need to do that with 1,000 profiles? If your already editing with the interface it really only takes a little time to go ahead and edit the name too. What did you do with prior units which only allowed 100 to 200 presets? I'm not judging, simply curious.

  • I do the same thing. Store the amps with different names in the device. Are you actually changing the tags or just the profile name?


    At the moment, just the names - but that's very limiting for me.
    I'm still getting a little weirdness i need to take care of with Laurent's tool (corrupt amp names), but once i get the hang of it i could, say, change the author of modified rigs to a song's name -
    then i could have a list of 60 or so basic profiles i'm using with their real details,
    and i could search by author to see all the rigs associated with a particular song to make set my MIDI parameters for a particular gig, which should save me a ton of time and trouble ('did i set the pre-chorus lower, or is the wrong song?' kind of thing)


    And as for your other question, when i worked simpler setups i'd compromise. I couldn't drop volume for certain parts of songs and couldn't have more than one delay setting (i worked analog before the KPA).
    The KPA is worth every penny, no doubt - but having dropped such an amount on a piece of digital gear, i don't expect to have to compromise anymore :thumbup:


    Even with (sort of) comparable products you'd have separate footswitching for every effect on every preset, which isn't available on the KPA at the moment.

    "But dignity is difficult to maintain
    stamina requires constant upkeep
    repetition is boring
    and you pay for grace."

    Edited once, last by Quitty ().