Notice: Google App Engine Conversion API to be decommissioned in November 2012

[Christina Ilvento]

In November, 2012, we will be decommissioning the Experimental Conversion API. In the 1.7.1 release, the API will be marked as “decommissioned” in our documentation, and starting the first release after November 1, the Conversion API will no longer be supported in the App Engine SDK or runtimes.

One of the great things about App Engine is our ability to quickly release and iterate on new Experimental features. But sometimes a feature doesn’t get the developer adoption and momentum needed to justify providing it as a fully supported feature with a continued maintenance commitment. The Conversion API is one of those features.

We are committed to providing a well-maintained, fully-supported platform and we will continue to focus on providing high value features for our developers. If you have any concerns or questions, please don’t hesitate to contact us at google-app-engine-conversion-api-decommission-questions@google.com.

Thank you for your continued support of App Engine.


Regards,
Christina Ilvento, Product Manager for the App Engine Conversion API

[Drake]

Crap. We had just planned to do some pretty big things with this… But it hadn’t happened because we were unable to bank on it being there forever…

 

How much adoption is needed to keep a feature?

 

Can I buy support for it?

 

 

--
You received this message because you are subscribed to the Google Groups "Google App Engine" group.
To post to this group, send email to google-a...@googlegroups.com.
To unsubscribe from this group, send email to google-appengi...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/google-appengine?hl=en.

[Benjamin Possolo]

Yeah this is unfortunate. I was planning on adding some features that used the OCR functions.

[Daniel Florey]

Can you please give an exact date when the API will be switched off?

We are using it in some of our apps and rely on it heavily - as I've not expected that this great feature will not make it into production some day.

We need to implement a proper replacement, so it would be great to get some more time.

What about keeping it running until the end of the year?

We are planning to implement a replacement running on Google Compute Engine. Hopefully this is not the next service to be doomed ;-)

Stay tuned.

Daniel

[doright]

I am also using this feature.  And in fact only recently put it into production, so it is very disappointing.

I understand that Google needs to test out lab features to gauge uptake. But it would be great if you could help us out in the following way:

1) give us more time to find an alternative, end of year would be much more reasonable I agree.

2) suggest alternatives ??!!

3) tell us how short you were from your required uptake.  This will help us make decisions in the future about what features to try out.

many thanks,

Doug

[Joshua Smith]

On Aug 27, 2012, at 5:20 AM, doright <doug.s...@gmail.com> wrote:

tell us how short you were from your required uptake

Although I never actually tried the conversion API (too many people complaining it didn't work well scared me off), I think that Google is setting a bad precedent here.

Killing projects because of low adoption makes people loath to adopt, because all the work they spend getting a feature integrated may be spontaneously wasted if it turns out the feature isn't widely adopted.

So people reasonably avoid adopting features until they are popular.

So features are killed because of low adoption.

etc.

I'm not sure exactly what the better way of doing things would be, but there are plenty of options. This way is clearly a terrible way to do business.

-Joshua

[Daniel Florey]

I suspect that the low adoption is the real reason behind it. I guess either the core developers left the App Engine team or they had some serious licensing trouble with the underlying products (PrinceXML etc.)

But this just a guess, it is fully understandable that Google will not communicate the real reasons behind this move.

But on the other hand I agree that claiming the "low adoption" is not a good excuse as there are too many services in experimental state and it will not scare developers to try them and invest time (and money).

So all I'm hoping for is to get some more time to find or implement an alternative and a little bit less of this marketing-driven email-communication.

If a service is switched off  I simply want to get an email that says "We are sorry..." instead of "One of the great things about App Engine..."

Daniel

[Christina Ilvento]

Hi All,

The API will be shut off in November of 2012. When we have an exact date for the release we'll let you know. 

Thanks,

Christina

--
 
 

[James Broberg]

This is a bit of a bummer. I was keen to try it out to generate PDF invoices for customers....

[Arne S]

I have added a request to reintroduce this functionality

http://code.google.com/p/googleappengine/issues/detail?id=8139 

On Friday, August 31, 2012 11:49:57 PM UTC+2, Bommelmutze wrote:

Does Google new provide any similar Library or API?
Actually it seems to be very hard to convert PDFs without the Conversion API to Images. Do you have any alternative way to do that?

[Jon Stevens]

I'll come up with something in October as I need this functionality. Basically, you just need a simple nodejs (or whatever) wrapper around imagemagick/ghostscript. I'll probably make it super simple to deploy a service to heroku.

jon

On Friday, August 31, 2012 2:49:57 PM UTC-7, Bommelmutze wrote:

Does Google new provide any similar Library or API?
Actually it seems to be very hard to convert PDFs without the Conversion API to Images. Do you have any alternative way to do that?

Am Dienstag, 21. August 2012 20:16:48 UTC+2 schrieb Christina Ilvento:

[Andrius A]

Hi All,

I am thinking to create a webservice for this as I have much of experience converting various file formats. 

How much would you be willing to pay per file conversion? per batch?

Regards,

Andrius

--
You received this message because you are subscribed to the Google Groups "Google App Engine" group.

To view this discussion on the web visit https://groups.google.com/d/msg/google-appengine/-/rjfSKRiAYXMJ.

[Jon Stevens]

I thought about making something that is worth charging for, but I think the market is too small. Thus, I think that this should be free. It should just be some simple webservice that people can deploy to heroku on a free tier app. It shouldn't be that hard to develop and if we all work together on it, we can make something nice. =)

jon

[Andrius A]

Just using imagemagick and ffmepg is easy, but it is not perfect in real life. It will fail occasionally with various PDF versions or different codecs, will leak memory or consume resources.

To make this work requires a lot of effort and using proprietary applications so it's not really viably to make it free.

[Jon Stevens]

Hey all,

I've integrated imagemagick/ghostscript into an easily deployed heroku/nodejs app (runs just fine on the free tier) that serves as a good enough replacement for the conversion api for me... check it out...

https://github.com/lookfirst/convert

best,

jon

On Monday, September 24, 2012 12:40:56 PM UTC-7, Stanislav Minar wrote:

what a pity. "But sometimes a feature doesn’t get the developer adoption" ... Don't you think it's because 100 conversions/day limit also for Billing enabled apps? https://developers.google.com/appengine/docs/quotas#Conversion

Just give us a possibility to buy conversions as we need.

S.

On Tuesday, August 21, 2012 8:16:48 PM UTC+2, Christina Ilvento wrote: