Re: APP WISH LIST: translation app, IFTTT

[kkiel]

Hi, is this the right place to look for someone who could write an app for me (paid)? The app should do the following:

1. The user types a term (a 'start term') in his main language into a start field (like 'König' into the headline field in German)
2. the app finds connected translations of the 'start term' in a user provided dictionary (a thesaurus?). For example: If the start term is 'König', the translation app finds 'king' and 'roi' in the dictionary
3. the app writes the translations into user defined fields (in my case: hijacked metadata-fields or else attribute fields)

I think of it as a simple 'if this then that' feature: If the start term in the source field (headline) is 'König' then the apps writes 'king' into the designated English term field and 'roi' into the designated French term field.

The user should be able to define the fields that are connected to a dictionary and the fields the translations are written into.

The dictionary should be able to able to hold 100000 terms at least.

Is that doable? Does an app like this already exist?

Thank you for your help!

kkiel

[Mario]

One solution for this, with built-in features, would be Attributes. Which are very powerful and unique to IMatch.

Some users manage multi-language data in per-file Attributes. For example, an Attribute Set with fields "lang" und "value". Value takes then "King", "König" and "roi".
To produce "output" in a certain language you can then use a Metadata Template to copy data from Attributes to keywords, title, description etc.
And you can always see the metadata in all languages in the Attributes panel.

A custom app like the one you have in mind would be more comfortable and doable.
I have written similar apps for commercial users in the past.

[Jingo]

I'm pretty sure an APP that reads a term, performs a lookup into a translation file and returns that translation to an attribute.  If you have a sample translation file, I'd be happy to see how much work would be involved to create an app.... let me know! - Andy.

[kkiel]

Thank you, Mario and Andy!

I first checked whether I missed a feature of the attributes functionality that allows me to somehow automate the process of inserting recurring translations into attributes.

In my case it's German, English, French and Latin names of species. Since there are hundreds of thousands of them, I would like to store a thesaurus/dictionary of all species I ever inserted into my database, type 'Rabe' into an attribute or metadata field and get 'raven', 'corbeau' and 'Corvus corax' written into the other designated fields automatically. Or chose 'Rabe' from a dropdown menu for the first field and get the translations into the other fields.
I would have to store 100000+ of such entries somehow. I could import these entries into a global attribute set, I guess, and edit them there, if necessary.

But how do I get the data stored in the global attribute set rows into the attribute rows of a single image file?

I've read the documentation again but I'm simply stuck with this. There has been a 'splash' functionality for attributes in a former version of IMatch that I have used extensively but at the momend I'm lost!

Thanks for your help!

[Mario]

This is not possible. At least not without a specialized app.

What you want is a built-in auto-translation with a database of sorts that holds your translations. And an app that manages all this.

Writing an app that does what you need is surely possible. But I doubt it is worth the effort only for just one user.
The app must be designed, written, tested, debugged. Kept up-to-date with changes in IMatch and maybe in your requirements in the future.
It will take a couple of days to design and develop. With several iterations etc. Not cheap, even if you hire a programmer via one of platforms like Fivver.

I know that some of the big (5,000+ US$ per year) DAM systems offer similar functionality, usually in in special (paid) add-ons for that purpose.
Maybe you should consider using one of these? IMatch can do a lot, but it cannot do everything the corporate-grade (and corporate price) systems can do.


There are a few IMatch users who work with multi-language content. Usually people who deal with scientific taxonomies of sorts and need to handle both the Latin and translated species name or something.
This is usually easily done in IMatch via the thesaurus and keywords and synonyms.
To store descriptions and similar in multiple languages, Attributes are a good way.  You can display the data, search it, sort by it, filter by it etc.
And you can later use a MD template to copy metadata in a selected language into the output file for publishing or clients.

XMP offers multi-language metadata for a few selected tags. But I have given up on them after learning that most other 'XMP-aware' software silently strips out multi-language data during import or save operations. Including some products from market leaders. There is simple no wide use for such features and the development efforts are not warranted or economical. So they all skip on it.

Besides, even having support for multi-lang XMP tags would not help in your case, because you want dynamic lookup of data you enter in your translation database.
I can see how this would help you with your special requirements. But I doubt that many other users will ever have a need for this. So I'm out.

[kkiel]

Hi Mario,

I understand and share your aversion against metadata in different languages. But a functionality that allows to comfortably translate recurring data from a user provided dictionary (not even on the fly with the help of deepL or google) is a feature that ALL picture agencies on earth need deperately. Picture agencies routinely sell in different countries (or send their images to bigger international agencies), therefore need to translate their metadata and pay through the nose for automatic data insertion. Though this is something that a database is predestined to do and, as you write, takes only a few days to build.
Why not disrupt the market of the big DAM providers with the great features of IMatch + a pro translation feature? And get those customers for IMatch (with a higher price for this 'agency version' but still significantly less expensive than the software they use know)? It should be easy to convince agencies to switch to IMatch then and I'd love to sell it to them. So I'm in!

[Mario]

Agencies usually have in-house systems for that. Or use one of the big DAM vendor systems.
Not much of a market there for IMatch.

The functionality you want is surely doable.

IMatch already uses machine translation provided by Google or DeepL in the App Translator and IMatch Translator apps, for example. Not many people know that.
See https://www.photools.com/imatch-translator/ and https://www.photools.com/translating-imatch-apps-app-translator/ for more information.

And, as always, I'm listening to my users.
My time is well spent with implementing features which are helpful for the majority of users.
Spending a days or a week or two with features which are massively helpful for only a handful of users is not.
Unless there is business behind that or a potential to sell a couple of hundred extra licenses.

So, I suggest you post a feature request. So we can see how many users have this kind of requirement.
If there are a sufficient amount of existing (or new users) who need such a thing, we can discuss it further after IMatch 2020 is out.

[kkiel]

Hi Mario,

that's great, thank you! I'll post it as a feature request. And thank you for pointing me to the App Translator and IMatch Translator apps, I didn't know them either. They will be great for my not-fixed-scientific metadata. And my offer to sell an agency version of IMatch still stands.

[kkiel]

ok, the two translation apps are for translation of IMatch itself, not for metadata

[Mario]

ok, the two translation apps are for translation of IMatch itself, not for metadata

Indeed. But a) translating metadata was never a requirement or feature request and b) the apps use Google/DeepL translation interfaces in the cloud. Which demonstrates how powerful apps in IMatch are.

[kkiel]

Yes, I know that IMatch is powerful and more so than several of the five-or-six-figure-DAM-programs. That why I would like to stick with it ...

[Jingo]

FYI.. I have created the backbone for an APP that seems to do what you are hoping for...  so far, the app will:

• Read a JSON translation file into an array by keyword/language translation
• Load an attribute schema for that includes each language
• Populate the attribute schema for selected files

I now need to take the user provided "term", search through the array and locate the term and then load the translations files from the array into the attribute schema.   The language file is in the format: {Original Term: English:<english>;Spanish;<spanish>;French:<french>; etc....}   The scheme also is setup to include the original term along with the translated terms. 

Works great so far... now, I will work on the array search/find and using that data to populate the attributes.

[Jingo]

App is all set!  The app now searches the array from the translation file, finds the term and applies the translations into the attribute set for each selected image.

I want to do a bit more refinement but should be fully complete shortly..  Here is a sample of what it looks like:




Translation JSON File:

[Jingo]

Well... not sure if the OP still needs this or not but I'm all done with the APP and figured I would share it!

As stated, the app will perform the following tasks:

•     Read a JSON translation file into an array by keyword/language translation
•     Load (and create if not already in existence) an attribute schema that includes each language
•     Populate the attribute schema for the selected files using the translation file/array

Once the app is in pace, the only thing the user needs to do is edit the trans.json file with the translations following the JSON format within.  You can add/remove languages by editing this file and also making the same edits to the code within the index.html file.  Of course, the code can also be modified to build attribute set for a variety of stuff... say, you would like to add specific attributes every time you encounter a flower image.... no problem!  Fill out the trans.json file with the KEY: Flower, add in the name/value attributes that should be added to each image, modify the code to match the name/value pairs and voila!  Now, every time you type "Flower" in the app, the attributes will be added to the selected images... 

It is very easy to modify the program/JSON file to auto-populate attributes with this app... I hope others can find value in it (and learn how to code IMatch APPs with the included code comments).

Here is the all important App Icon:


And the Final screenshots:


Translation JSON File:



Mario - is it possible for you to edit the title of this thread to remove APP WISH and change to APP: or do I need to copy/paste this into a new thread?

To install, just unzip the Translator.zip file and place the app folder into your C:\ProgramData\photools.com\imatch6\webroot\user\ path.  When you launch IMatch, the app will appear in the app manager.  As always.. use with caution... and be sure you know what you are doing! 

Enjoy!!

[kkiel]

Hi Jingo,
I found your replies only today after I had a major motherboard crash - thank you very much!! I still have to work a little bit to re-install IMatch on another PC temporarily but will try your app asap.
Regards,
kkiel

[Jingo]

Hi Jingo,
I found your replies only today after I had a major motherboard crash - thank you very much!! I still have to work a little bit to re-install IMatch on another PC temporarily but will try your app asap.
Regards,
kkiel

My pleasure kkiel.... I hope the app provides you with the functionality you were looking for (or close to it).  If I can help further.. please let me know!  Thx - Andy.

[kkiel]

Hi Jingo,

so I finally restored my workspace so that I was able to install your app into a copy of my IMatch database (thank you very much again for your efforts!!). I manipulated the trans.json and the index.html files so they contain an example of my wanted translations and my wanted languages. Though, when I try to insert a word into your app interface (like 'Brot') nothing happens. There is no view like in your screenshot saying 'Found and set the following image attributes'.

Do I have to do something else?

And, after this is solved: How exactly do I get the translations into their destination fields? Put the cursor into the destination field and click onto the app provided translation?

Thank you very much for your help!

kkiel

[Jingo]

Hi Jingo,

so I finally restored my workspace so that I was able to install your app into a copy of my IMatch database (thank you very much again for your efforts!!). I manipulated the trans.json and the index.html files so they contain an example of my wanted translations and my wanted languages. Though, when I try to insert a word into your app interface (like 'Brot') nothing happens. There is no view like in your screenshot saying 'Found and set the following image attributes'.

Do I have to do something else?

And, after this is solved: How exactly do I get the translations into their destination fields? Put the cursor into the destination field and click onto the app provided translation?

Thank you very much for your help!

kkiel

Hi Kkiel - no worries... if you can post the 2 modified files here, I'll take a look just to be sure the correct setup occurred.  JS is robust - but finickey and there could be a syntax error to the index.html file.  I would suggest trying the original version just to be sure it is working 100% as well.  Thx! - Andy.

[kkiel]

Hi Jingo, thank you!

So I started with your fresh files and this time it almost worked: I found that the app creates a 'translate' attribute set with the columns Original, English, Spanisch, French, good! When I chose 'Brot' it shows the list 'Found and set the following image attributes' with the deposited translations, good!
But this doesn't work with the second term 'König', though. It doesn't show the list then.

Whether I try it with Brot oder König - the attribute rows aren't filled with the translations, they stay empty.

Maybe there is some default setting in my installation of IMatch that prevents autofill? But I have no idea what this can be.

Regards!

kkiel

[Jingo]

Hi again.. great.. I'm glad it worked out of the box so that means something might have gone wrong with the edits you made to the files... if you can post them here, I'm happy to take a peek.

Curious if you tried "konig" in the search box without the umlaut?  I built it without in the JSON file so I believe that may be why it is not working for you.... it will need to be added in the JSON file for it to be found as 'König' (I'm doing a straight compare so diatrics need to be included in the translation file AND in the search box).

Hope this helps! - Andy.

[thrinn]

Andy,
I think the problem is that you always use setid: 1. This variable is set in line 143. And while you are retrieving the Attribute scheme in line 158, the POST statement in line 186 still uses the fixed value.
If the Translate set is not the first attribute set in the database, one gets a "Bad Request" error because the app tries to write to the first attribute set.
I just added the last two lines to overwrite params.setid with the retrieved ID - and voila!

Code Select Expand


          IMWS.get('v1/attributes/schema', {
            set: 'Translate'
          }).then(function(response) {
            schema = response.schema;
            // Use the ID of Translate attribute set
            params.setid = schema[0].id;
...


Regarding Umlauts: I added an entry König to the JSON - and this also works as expected!

Code Select Expand

         
{
  "Konig": {
    "English": "King",
    "Spanish": "Real",
    "French": "Roi"
  },
  "König": {
    "English": "King",
    "Spanish": "Real",
    "French": "Roi"
  },
  "Brot": {
    "English": "Bread",
    "Spanish": "Pan",
    "French": "Pain"
  }
}


[Jingo]

Andy,
I think the problem is that you always use setid: 1. This variable is set in line 143. And while you are retrieving the Attribute scheme in line 158, the POST statement in line 186 still uses the fixed value.
If the Translate set is not the first attribute set in the database, one gets a "Bad Request" error because the app tries to write to the first attribute set.
I just added the last two lines to overwrite params.setid with the retrieved ID - and voila!

Code Select Expand


          IMWS.get('v1/attributes/schema', {
            set: 'Translate'
          }).then(function(response) {
            schema = response.schema;
            // Use the ID of Translate attribute set
            params.setid = schema[0].id;
...


Thanks Thorston... Yup... my mistake on that one... since I don't use attributes - I only had a single schema thus I hard coded 1 at the start....  and it always worked!   Attached, please use this update zip file.  Thx!

[kkiel]

Hi Andy, hi Thorsten,

thank you very much for working on this app! And I'm very sorry that it takes me so long to test and reply, I still don't have my regular PC back from repair (Cyberport anybody?!).
I tried the new version: now I get the 'Brot'-translations correctly, but 'König' still doesn't work. 'Konig' works, but it enters three rows of translation attributes ...

Thanks a lot again,
regards,
kkiel

[Jingo]

Hi Andy, hi Thorsten,

thank you very much for working on this app! And I'm very sorry that it takes me so long to test and reply, I still don't have my regular PC back from repair (Cyberport anybody?!).
I tried the new version: now I get the 'Brot'-translations correctly, but 'König' still doesn't work. 'Konig' works, but it enters three rows of translation attributes ...

Thanks a lot again,
regards,
kkiel

Hi Kkiel - my pleasure!  Can you please show a few screenshots of the attributes panel for an image and a screenshot of your Translation.JSON file?  I just tried this again on a clean install - works perfectly once I modify the JSON file to:

Code Select Expand

{
  "König": {
    "English": "King",
    "Spanish": "Real",
    "French": "Roi"
  },
  "Brot": {
    "English": "Bread",
    "Spanish": "Pan",
    "French": "Pain"
  }

[kkiel]

Hi Jingo,

here's a screenshot: 'Brot' worked, 'König' not.
The translation.json-file is exactly the one I downloaded from here, no changes made.
This is the content:

{
    "Konig":{
        "English":"King",
        "Spanish":"Real",
        "French":"Roi"
    },
    "Brot":{
        "English":"Bread",
        "Spanish":"Pan",
        "French":"Pain"
    }
}

Thank you very much!
kkiel

[Mario]

Should'nt

"Konig":{

be

"König":{

[kkiel]

And, I forgot: If I type 'Konig', three rows are made, see screenshot.
Thank you!!

[kkiel]

And, if I correct Konig to König in the trans.json-file and type 'König'  ... it works perfectly. Ok! This doesn't explain why it types Konig three times but it's great now!

[thrinn]

This doesn't explain why it types Konig three times but it's great now!

Thr app does not check if a corresponding attribute record is already present. If you use it multiple times on the same picture, one record is added every time.

[kkiel]

Yes, I guess I have pressed the button several times ... Sorry for thinking it was the app! I will test it with some 'real' English, French and Scientific names now and provide a source file (trans.json) with some translations if anybody is interested. But this will take some time.
I really am thankful for everybody helping me with this issue!!

[Jingo]

Glad it works for you and thx for verifying things are correct.  Yes, I didn't add any checking to the app to prevent multiple instances ... though this is something I could look into doing (would need a new checkbox (overwrite/append), get attributes for selected files, update ID list and overwrite for empty schema (overwrite) or file selected file list (append).  Something to think about.... glad the app is working though! - Andy.

[kkiel]

I just wanted to follow up with this thread since I have been able to install Jingo's translate app in my standard working environment after I got my PC back. It works great and spares me hours of work every day. This app really is a great time saver for everybody who has to fill the same fields with translations (or other content) over and over again. Finally no more copy and paste! Thank you very very much again, Jingo!

Regards,

kkiel

[Jingo]

I just wanted to follow up with this thread since I have been able to install Jingo's translate app in my standard working environment after I got my PC back. It works great and spares me hours of work every day. This app really is a great time saver for everybody who has to fill the same fields with translations (or other content) over and over again. Finally no more copy and paste! Thank you very very much again, Jingo!

Regards,

kkiel

My Pleasure kkiel.... this is what the community is all about... and my small way of giving back! 

fyi - in the 2020 version, duplicates will no longer appear because the IMWS function has been updated to restrict based on the unique flag (https://www.photools.com/community/index.php?topic=9438.0) - just something to look forward to down the road!

Enjoy the day..  Andy.

[kkiel]

Hi Andy, hi Mario,

I've been working on my metadata translation for some time now with Jingo's great app and I think this solution provides way more than simply helping me populate some fields. I can now store whole sets of metadata and manipulate them outside of IMatch without having to link them to to specific image files and import them via csv. This is huge. Maybe you programmers did realize this before but I as a non-programmer did not. And maybe I'm overlooking some possiblity to do this in IMatch itself.

By now my workflow for manipulating metadata now is as follows:

Open IMatch, open Jingo's trans.json-file in an editor, add and manipulate metadata in the trans.json-file, hit save trans.json-file, chose an image in IMatch, type the focus keyword into the Jingo-app, and whoosh!, the complete metadata row for a special focus keyword gets written into my IMatch file attributes. This works very fast - much faster than reading the sentences above.

This is an example of my data input:

    "Zuckerwurzel":{
        "DeutscherName":"Zuckerwurzel, Zuckerwurz",
        "DeutscheBeschreibung":"Zuckerwurzel",
        "DeutscheKeywords":"Zuckerwurzel; Zuckerwurz",
        "EnglischerName":"skirret",
        "EnglischeKeywords":"skirret",
        "FranzoesischerName":"chervis",
        "FranzoesischeKeywords":"chervis",
        "LateinischerName":"Sium sisarum"
    },
    "Linné":{
        "DeutscherName":"Carl von Linné",
        "DeutscheBeschreibung":"Carl von Linné (1707-1778), schwedischer Naturforscher",
        "DeutscheKeywords":"Linné; Botaniker; Carl von Linné; Naturforscher; Botanik",
        "EnglischerName":"Carl Linnaeus",
        "EnglischeKeywords":"Carl Linnaeus; Linnaeus; botanist; Carl von Linné; naturalist; botany",
        "FranzoesischerName":"Carl von Linné",
        "FranzoesischeKeywords":"Linné; botaniste; Carl von Linné; naturaliste; botanique",
        "LateinischerName":""
    },

Once written into the trans.json-file I never have to copy and paste data into the attribute rows again and I'm very happy with that. I am building my very own multilingual controlled vocabulary tool - great!

And I'm planning to use the app for way more than translations, too, like inserting data about the source of each file etc.

At the moment I'm searching for a solution to comfortably manipulate the trans.json-file within a grid so that it's alphabetically sortable (to answer questions like: 'Did I already enter 'Mais'?!) but it looks like there are some solutions out there that I still have to test.

It would be great to store and manipulate this information in an attribute set in IMatch itself, then, when adding data to a referenced attribute row, being able to choose between inserting 'reference only' or "full row of metadata". This should be quite as elegant. But I image it could get labour-intensive when data in an attribute set should be searchable, sortable etc. And maybe with my envisioned hundreds of thousands of entries I could blast the scope of IMatch/Windows.

Anyway: Thank you both very much again for your reliable tools! I'd be happy to provide my 'vocabulary' trans.json file (containing crop plants first) here at the forum to give sth. back, too, if anybody's interested.

Regards,

kkiel

[Jingo]

Hi kkiel... thanks for the kind words... very happy to hear the app is proving useful to you - the BEST news a programmer can hear in fact!

Check out this simple online tool to sort the contents of the JSON file... it should provide you with the functionality you are looking for to quickly find duplicates: https://novicelab.org/jsonabc - just copy/paste the contents of the JSON file and voila.... an instant alpha sort!

Would love to see the completed JSON file when finished.... and also - would love to see any manipulations to the software you make to use this APP in a different/unique way....

Enjoy! - Andy.

[Mario]

What you do here is rather special.

Reference attributes allow to you refer to globally stored data without duplication.
Attributes make maintaining multi-language keyword and contents very easy.
Metadata Templates allow you to fill in tons of metadata with a single mouse click (via Favorites). And to use multi-language Attribute data to fill metadata.
The Thesaurus allows you to insert individual keywords, any number of synonyms or even entire hierarchical trees of keywords with a single click. In one or more languages.
IMatch really offers a lot in this area. More than most users will ever need.

Doing on-demand translations of metadata values from a centrally maintained vocabulary is not something many 'normal' users need. This is more common for stock photographers and some of the commercial/institutional IMatch users.  But these often have solutions in place already.

If more users would need this, it would show up regularly in the feature request board or via support emails.

You have now a free purpose-built app which implements exactly your personal workflow. Thanks to Jingo.
Doing the same in a more flexible fashion suitable/adaptable for a larger user audience would be doable. Or even as a built-in feature. But I doubt that it would pay off.
If suddenly a couple of hundred stock photographers decide to leave the Adobe cloud and switch to IMatch, under the condition that I would create such an app/extension, I would give it some real thought. But not before.

[sinus]

...
You have now a free purpose-built app which implements exactly your personal workflow. Thanks to Jingo.

I agree fully.
Jingo (Andy) is a nice and very helpful person (others too, of coourse, like Thorsten, who has helped her also).

There is a difference when someone does something, whether he earns his living with it or whether he does it out of joy, quasi as a hobby.
If you "have to" live from it, then I understand that you have to consider whether it is worth it or not.
But if you do this as a hobby, then you don't necessarily have to see whether it's financially worth it.

And Andy has proven many times that he wants to help without any financial advantage.
And I think that's very nice and cool from him.

But if you have to make a living from it, like Mario, then it's understandable that you always have to weigh up how many people are interested and whether it's worth it.

In any case, the threader (in this case kkiel) has received an outstanding app that obviously helps him a lot and that he appreciates.

That's the beauty of this community, if possible, users help other users.
Thank you all for that.

[thrinn]

That's the beauty of this community, if possible, users help other users.

A friendly community is very valuable indeed!
And not to forget: Which other product than IMatch allows us the flexibility to add such very special solutions!

[Jingo]

Agree all around... great product by Mario and a wonderful community... I'm always happy to give back in my small way. 

[kkiel]

Hi Jingo, Mario, Sinus and Thorsten,

thank you very much for your replies and efforts! You're right, I got a very helpful tool for my special workflow for free (though I offered to pay somebody to implement something like this in my first post). That's amazing and I'm very grateful to Jingo and the community for this.

And I indeed have a one-woman-stock-agency (that's still a side projekt and not yet profitable, by the way) with very special needs and enormous amounts of metadata. And I unfortunately am no programmer and struggle a lot with the thinking of those who are. I stick to IMatch since 2002 (!) when a developer friend recommended it to me. I consider myself a 'power user' of IMatch, as measured by the thousands upon thousands of hours I spent with it. I always wonder how anybody would favour the adobe DAM tools that lock you in a monthly payment agreement and still can't do some things that IMatch can, like auto-categorize images with a special tag. Or provide a clean solution for multi-language metadata storage. I know that some of the really expensive DAM solutions (five figures and up) offer things like that. But I've also learned that they completely and deliberately neglect the stock agency's und photographer's market because huge businesses with a lot of marketing material are a more convenient and more lucrative market for them.

So I somehow have an existential interest in IMatch opening up to the needs of small stock agencies and photographers who sell their material internationally (and who can afford to not market their material internationally?!).

I understand Mario perfectly for not complying to any software wish of his users. And to base his decisions on rentability. Of course, I hope he does! But I think there should and could be a much bigger market for IMatch. Photoshop sees serious competition from tools like affinity or luminar. Why not take away customers from lightroom who only grudgingly stick to their subscription model? And offer multilanguage metadata support that is, as far as I can see, the most important feature a stock agency needs?

IMatch is great with it's reliability, it's abundancy of features and it's great support and community. It does WAY more than most of the other DAM solutions. So why not offer a 'pro' version for 10 times it's price or more? I would happily pay for that. If I could only buy it.

Mario, please don't read this as criticism. It is the opposite of that. I am absolutely amazed by a software solution that is so great and so modest at the same time.

Regards,

kkiel

[Mario]

Mario, please don't read this as criticism. It is the opposite of that. I am absolutely amazed by a software solution that is so great and so modest at the same time.

I don't. I welcome ideas, feedback and feature ideas from users!
If you think that what this special app does for you would be helpful for more than a handful of users, write up a feature request in the corresponding board. Other users can see it there and discuss it. Not many users participate in this community and only a few of them read the posts in the scripting board. The feature request board has much better chances.

But I think there should and could be a much bigger market for IMatch. Photoshop sees serious competition from tools like affinity or luminar. Why not take away customers from lightroom who only grudgingly stick to their subscription model?

I would love to sell thousands of extra licenses each year! Let me know your ideas.

Quite a number of people has discovered and moved to IMatch after Lightroom went subscription-only. That's why there is a Lr importer app in IMatch 
Lightroom is not a bad product if you can live with rather limited DAM features (many people do, because they don't know better) and a (probably) life-long subscription, it is a valid choice.
Not the best (if you combine IMatch with a another RAW developer you may get better images and surely get a lot better DAM).
Affinity products are also good. They are not as good as Photoshop or InDesign - but they are much cheaper and require no subscriptions (yet).
And cheaper appeals to most users out there. Me included. I use Affinity Designer, not InDesign. But I use Photoshop because I need what it does.

IMatch is a niche product. I would be happy to get suggestions for how to make it more prominent. Let me know if you have an idea.
Adobe spend 150 million dollars (!) each year on marketing and the big DAM vendors also have much deeper pockets than I have.
I cannot afford marketing. For example, the minimal monthly budget for Google ads to make some impact is about 2,000US$. Better is 5,000 to 10,000US$. That's not much for a startup or a 'real' company, but totally out of reach for me...

Stock is a tough business and it has become a pure mass business over the past years. Even Getty Images (!) will soon stop selling royalty-based images and switching to the royalty free mass image business. Free stock sites like Pixabay or Unslpash offer millions of free photos. No business there.
Stock photo sites like CanStockPhoto sell royalty-free images for a few cents each.

Implementing special stock photography features is certainly doable. But I wonder how many new users this would bring?
All stock photo sites I know offer fairly usable upload facilities/tools for this purpose.