Instapaper now requires a login

Are updates in progress for both DTPO and DTTG?

Wow, that’s seriously scary.

As a product owner myself, I have to ask how it is that you guys not have a predictable release schedule or even a rough timeline for delivery? Does this mean you’ve decided to build the feature but haven’t started on it, which will break my workflow for months or potentially longer?

I hate to switch back to Evernote after dropping a bunch of money on you guys, but it’s starting to feel like that’s what I’m going to have to do.

As noted here: Instapaper now requires a login We are working on it for the next maintenance release. It is our policy that we do not issue timeframes for our and our customers’ sakes. That says nothing about whether we are working on something or not, and we’ve already said we are working on it in this case.

It’s already working over here but just needs some more testing. An update should be available later this month.

I have to say that this feels to me like a major over-reaction. Keeping quiet about release schedules is (in my experience) a common practice to avoid problems if, for any reason (and there are plenty) the schedule is missed.

On the other hand, responding within days to a reported problem is decidedly not common practice. If the fix is, essentially, to replace Instapaper service with something in-house, it’s not trivial.

It seems to me that the DT guys can’t win - no release date and back to Evernote; publish a release date and maybe get slammed because it’s not soon enough; publish and miss a release date and get slammed for letting customers down. Especially frustrating since DT isn’t the cause of the problem - something has changed at the Instapaper end.

You might consider it an overreaction, I think my workflow was broken by a cloud failure that’s going to take weeks to resolve. It may not be important as a feature to everyone, but it matters to me. I would not have switched to DT without that feature.

That aside I think the core problem really is that:

• DT designed and built a tool that people use to store important data and build into their workflows. This is treated just like any other desktop software, however…
• By including Instapaper into that workflow DT has also become a cloud services provider, albeit indirectly.

If they had stuck to the first model, and this was a new feature, I would have no problem waiting. I can’t even tell you how many years I’ve been waiting for Omnifocus to build multiple contexts into the tool. You could even tell me not to upgrade my OS with desktop software until you’ve got time to test, and that works too because if I upgrade and it breaks, that’s on me.

However, what you have right now is a service outage. It’s not one of DTs making, but it’s one that is occurring none the less because you integrated a cloud service into the product. Clearly that integration did not have proper lines of communication with the third party that is being relied on for this feature. Clearly there was no solid business relationship as well or they would have warned you and / or allowed a temporary work around.

What you need to do until this is resolved is set aside some of your regular operating procedures and act to resolve the outage. Failure to do so only hurts your brand further. I can’t recommend anyone use this software given the way this is going, and I get asked a lot about how I work for various reasons.

I think you guys need to re-evaluate your stances on how you provide information to users and resolve issues.

Harsh? You’re making a lot of assumptions about who knew and did what. Alternative facts, I suppose. Instapaper recently had a massive database failure because it didn’t understand how to use the services it bought from AWS. It seems DEVONtech is reacting prudently to the continuing decline of Instapaper as a viable, trustworthy service. Meanwhile, there are numerous other methods to saving documents in ad-free format from a browser to DEVONthink.

Sorry I though this was a forum to discuss the product openly. Am I only supposed to post positive things?

Yes, Instapaper’s outage was pretty bad, it’s clearly not a reliable service. Integrating that into your product is certainly a questionable choice these days with that info, but that’s not particularly relevant as that occurred after the integration.

However that should have probably been enough of a trigger to start the move, and maybe it was. If we are getting the fix this month either DT has been working on it or it wasn’t really that complicated a problem in the first place.

The lack of communication on what exactly happened and why is what surprises me. Given that Jim clearly wasn’t aware that something had changed tells me there isn’t a strong relationship. That’s not a dig at Jim, it just is what occurred. If there was we might have even got an early heads up it was going to happen, because in a good relationship you tell your partners about changes first and let them make plans.

And you are correct, there are other services that are not integrated into their web clipper. I can now use one of them and go through the extra steps of saving it as a PDF, importing that into DT, which is just more friction in the system.

Maybe I’m just wrong in assuming that I can rely on the software and services I pay for to perform the functions listed on the site.

Open means open - it means people can disagree with you, openly, which is what I’m doing.

I think your logic is skewed - if DT relying on Instapaper is questionable, so is your relying on it in your workflow. If Instapaper is unreliable because of an AWS outage (if that was the cause), every other AWS-based service is unreliable.

You’ve piled up a number of assumptions to conclude that either DT started the transition a while ago or the transition is simple; such evidence as there is (the responses in this thread alongside the thread timeline) doesn’t support that conclusion. To be clear, that doesn’t mean you’re incorrect; it just means the evidence for it isn’t visible.

Feel free to express a different opinion; don’t feel free to complain that the discussion isn’t open.

Add me to the list of impaired users! This Instapaper login thing has really interrupted my workflow. Looking forward to the DT update as soon as possible, please.

– Robert

Just adding my name to the list to follow the topic.


The following workaround may be of use to those who can’t wait until the next release:


Capture page:

  • send the page you would like to capture from your browser to Instapaper (by means of the Instapaper extension or the Instapaper bookmarklet)
  • open the Instapaper bookmark in DEVONThink (log in if you aren’t)
  • click on the link just added
  • capture the resulting page in your preferred format by means of the contextual menu or by choosing Data → Capture → [format] in the main menu

Be aware that in this way DEVONThink records Instapaper’s URL, rather than the original one (as the clipper does). You might want to change this by hand.

Looks like DevonThink 2.9.11 fixes this and internalizes the de-cluttering. It’s working awesome for me.

@jeff: Glad to hear it! :smiley:

That’s cool that the Mac version is updated, but what about DTTG?

Also, it would be nice if it could preserve images that are part of an article. So far, at least from what I’ve used it on, it doesn’t seem to.

DTTG2 will be updated when ready.

Thanks - less than 2 weeks from the first report of a problem to an update.

@tamaracks: Some articles do seem to retain the images while others do not. My unscientific testing suggests that the images not included do not have normal image HTML tags, and are either programmatically inserted or use a binary reference. Often, Instapaper and a browser’s reader mode also omit those same images. (The latter is a great way to check whether it is a DT “clutter-free” issue or a site limitation.) That’s not always the case, but a high percentage of my tests have pointed in that direction.

@ThatGuy: Agree 100%. Thanks DEVON folks!

Can we keep the Readability service as an alternative option?

The Readability service is dead.