DeepMeta - Manage your iStock portfolioDeepMeta - Manage your iStock portfolio Support [ ]
Articles- DeepMeta 3
Introducing DeepMeta 3 and the iStock/Getty unification
DEEPMETA 3 - Now available!

View the quick introduction video of the functionality and workflow.

Before contacting support, please read through the F.A.Q. (Frequently Asked Questions) section.

2017 marks the 10th anniversary of DeepMeta and we're celebrating this with a brand new version. DeepMeta 3 will be released on Feb 1st 2017 for macOS and Windows. This release also coincides with the iStock/Getty unification.

In short

If you don't have time to read the entire article, here are some important points to take note of:

  • On Feb 1st, all current 2.x versions of DeepMeta will loose the ability to fetch data from the iStock servers and uploads will no longer be possible.
  • DeepMeta v3.0 will be available for download, starting Feb 1st, depending on when iStock/Getty enabled the required server access. Once this is done, we will do some final tests and make V3 available for download.
  • This new version will communicate with new servers hosted by Getty, using a completely new API (interface used by computers to talk to servers).
  • After installing DeepMeta 3, you can still run the old DeepMeta v2.x version to check the (old) statistics already fetched in DeepMeta. You can also choose to uninstall the old version if you prefer.
  • DeepMeta 3 will allow you to manage and upload the following types: Images, Illustrations (raster & vector), Video and Editorial Images.
  • IMPORTANT: Your current data in DeepMeta will not be transferred to DeepMeta 3, which will start up empty! Until Feb 1st, you are advised to limit the number of non-uploaded files in DeepMeta to a minimum!
  • In the weeks following the launch of Feb 1st, I will be adding an import tool to DeepMeta 3, allowing to import selected data from the old version, including non-uploaded files. Note however, that the new keyword vocabulary will be different, so your term-selections for keywords will not be imported.

The iStock/Getty unification

Please refer to this article (link requires login) for all official info regarding the iStock/Getty unification. I will only address issues that affect DeepMeta.

DeepMeta is stand-alone software, running locally on your computer. All information you enter in DeepMeta is stored locally on your hard disk. To exchange data with iStock, DeepMeta does not communicate with the www.istock.com website itself, but uses an "API" on the iStock servers. This is an interface for computer-to-computer data exchange.

As part of the iStock/Getty unification, starting Feb 1st, the old iStock API will be retired, and DeepMeta 3 will instead talk to a new API, hosted on Getty servers. iStock/Getty call this the "EPS" (Enterprise Submission Platform). This is great news, since the new API is based on standard cloud technology, which has very high (upload) performance and is very stable. Finally, DeepMeta will now finally also be able to upload video files!

The way data is exchanged is completely different from what it was before, which means this part of the DeepMeta software had to be rewritten.

The iStock/Getty development team has been working very hard to create the new server system in the background. You may have noticed this resulted in some degraded functionality in the current/old iStock site, with very limited statistics at the moment, also in DeepMeta. I suppose this is the price to pay, which should soon be offset by the advantages of a reliable new system.

DeepMeta 3

Since so many things were going to change in the way DeepMeta exchanges data with the iStock/Getty servers, I've decided to take the opportunity to give the internals of DeepMeta a complete overhaul.

Long-term users will know that many features have come and gone in DeepMeta's 10 year existence: ratings, comments, ever-changing file types such as audio, logo and flash ... The development framework I was using to be able to create both a Windows and Mac version was also no longer actively supported by its manufacturers.

This all means that DeepMeta 3 has a lot of old code cleaned up, and lots of new code instead. The UI has also been recreated from scratch. We're ready for the next 10 years! I'm hoping to keep the UI and functionality differences between the Windows and Mac version very limited from now on.

Statistics

DeepMeta v2.x provided statistics on file-per-file basis, for things such as number of views, downloads, as well as royalties earned. The initial version of DeepMeta 3 will have none of those, for the simple reason that the API functionality on the new iStock/Getty servers for fetching statistics will not be ready by Feb 1st. You will however be able to access certain statistics on the ESP website starting Feb 1st. It just won't be in DeepMeta yet.

As for which exact statistics will come to DeepMeta, things are still a little unclear. From the info I have so far, we will get download (and maybe view) stats per individual download per date. Fincancial data such as royalties will not come to DeepMeta however, at least not in short term.

OS support

DeepMeta 3 will support macOS versions 10.7 (Lion) up to 10.12 (Sierra) and Windows Vista up to Windows 10. For a few years already, Windows XP was no longer supported, but still worked on XP.

Starting with DeepMeta 3, it will no longer be possible to run on Windows XP. This choice is dictated by the development tools I use. There's nothing I can change about this, unfortunately. The same goes for macOS versions 10.6 (Snow Leopard) and older: DeepMeta 3 will not run on these.

Local DeepMeta data

DeepMeta 3 uses a completely new library for storing its local data, and will show empty lists upon first startup. Any data you had in DeepMeta v2.x (including prepared files you have not uploaded yet) is still safely stored on your hard disk, but is not loaded in DeepMeta 3, since it's not compatible with this new version.

In the weeks that follow the launch of DeepMeta 3 of Feb 1st, I will create an import feature that will allow you to import certain older data into DeepMeta 3. This will include prepared files you have not yet uploaded, along with their title, description, releases and keywords. Please note that any disambiguation (ticking the term boxes for each keyword) will be lost. This is simply because the new iStock/Getty API uses a different vocabulary.

In any case, I strongly suggest to keep the number of non-uploaded files in DeepMeta to a minimum in this phase. Upload what you can and "empty" DeepMeta as much as possible.

User Interface changes

If you're already using DeepMeta, you'll feel right at home in DeepMeta 3. There are the usual Files and Releases lists, the Upload section and Preferences/Options dialog. A new concept is the Submission Batch. The new iStock/Getty ESP upload system will demand every uploaded file to be part of a "Batch".

A Batch is a group of related files of the same type (Image, Illustration, Video or Editorial Image). You cannot put e.g. Images and Videos in the same batch, even if they are from the same shoot. You will also notice some extra info that goes with each type of data. E.g. Model releases now require a birthdate and gender to be valid.

Tutorial & User guide

I will update the "Getting started" user guide here at this site when DeepMeta 3 will go live, explaining all the changes.

Keyword Wizard

This online tool (top right menu option on this page) will go away on Feb 1st. Its use and popularity was limited, and now with the change of vocabulary from iStock to Getty, would no longer yield correct results.

I'm planning an improved system for keyword suggestions as built in feature of the DeepMeta software. Deep learning technology and advances in image recognition technology offer some great opportunities.

Price

DeepMeta will remain completely free to all iStock contributors. This is made possible by an agreement where iStock/Getty presents DeepMeta as its officially endorsed upload and portfolio management tool to all its contributors.

Questions ?

If you have any DeepMeta related questions: as always, feel free to ask them directly to me, or post them on the DeepMeta support forum thread (link requires login), which I closely monitor as well. I usually respond within 48 hours.