Hi Dan, thanks for taking the time. We have not heard from anybody from Google in this forum for many months so your reassuring communication is very welcome.
I list below some of the reasons that might explain why some of us who have been following GAE for a long time have been skeptical about Google’s commitment and jump to conclusion when articles like this appears in the press. 1. There are still 3,000 issues open in the tracker. Although many of them are irrelevant enhancements others are critical. For instance, ---we never expected that 5 years later would be still unable to send 8-bit e-mail through the platform (issue 2383) or —in the critical area of security we would have to deal with a Users API stuck in the 2009 reality and crashes in certain important use cases (issues 9045 and 8916) or --- that it would still be so difficult to create an SSL app (issue 8528). 2. Now what makes this more frustrating is that bugs have been aging for way too long. For instance, issue 2383 was filed in 2009, it was accepted in 2012 and at the end of 2014 it is still open. 3. Google used to communicate a roadmap here. This was great for our planning. At some point the roadmap disappeared, again without communication. Furthermore, until a few months ago new releases (and pre-releases) were announced in this forum. Then suddenly the announcements and every Googler disappeared without warning. StackOverflow is a great Q&A forum but is not a discussion or announcement forum. 4. Finally, last but not least the enthusiasm on Google’s part does not come across. I have been following GAE and developing since almost day one. The early days the developers were out here and in the irc chatrooms all the time. Input from customers was actively sought. Now we do not see anybody from your engineering/PM team here in the trenches. Knowing the alternatives, I remain very enthusiastic about PaaS in general and GAE in particular. I acknowledge that the GAE stability is very good, Google’s innovations in the hybrid IaaS/PaaS cloud is significant and that recently I have seen more activity in the issues tracker. For instance, I was impressed how proactive you were when I filed 11396 or regression 10503. I remain optimistic that the GAE stability will stay where it is but also that the Google investment reflected in faster feature velocity will increase, and the open communication will return. Best, PK http://www.gae123.com On November 10, 2014 at 8:09:09 PM, Daniel Sturman ([email protected]) wrote: Hey fellow App Engine users, There is some great conversation in this thread. I’ll try to address some of the key points being discussed. Regarding the discussion group; our apologies for the delayed response. Most of our customer questions now come on Stack Overflow, so we’ve been monitoring it more actively than this forum. We’ll be watching this forum more closely too from now on. Regarding the larger topic of Google’s investment in App Engine: App Engine is a critical part of our cloud story, and will continue to be. We’re investing heavily in it. In the most recent months this investment has had two major prongs - stability improvements and new efforts to create a more flexible model within App Engine. First, stability improvements. App Engine has grown and so has the size and sophistication of the workloads that relied on it (thanks to developers like you). We realized it was time to take a step back and invest in driving down technical debt and improving overall stability as a foundation for the future. The team has been heads down improving stability and reliability. Some of the improvements include more comprehensive monitoring across all services, better application scheduling and load balancing, deployment of SSD to reduce latency variability for Datastore access, and many others large and small. Second, a more flexible PaaS. App Engine’s prescriptive environment for building web and mobile applications allows teams to iterate quickly on new ideas and scale up the ones that stick. The drawback, though, comes in terms of its constraints (e.g. limited JRE access, limited C/C++ Python modules, no inbound socket support). When we were building out our IaaS offering, Compute Engine, we realized that by unifying the compute stack (layering App Engine on Compute Engine), we could continue to give our customers the developer experience and efficiencies that App Engine brings with the flexibility and power that’s normally only associated with IaaS. Further, since it is a single stack, users can drop down into the IaaS layers when needed to make lower-level customizations (although we hope that most will never have to). We’ve surfaced all of this work as App Engine Managed VMs, which are now in Beta and open to everyone that wants a test drive. You’ll see that Managed VMs do not require you to manage the OS or web server configuration, and frontend serving has all the same great features as our existing runtimes. In other words, they marry the best of App Engine with a more flexible application environment. Finally, unified administration tools are an important part of a cohesive platform. This is the goal of the Developers Console. In some cases the cutover has been a straight “drop in” of existing functionality, in others we took the opportunity to make improvements. Not all is perfect, so thank you for the feedback! I’ve created bugs / feature requests for the items you’ve mentioned (infinite scroll issues, “save as” issues, better Task Queue admin functionality) and suggest that any other feedback be sent to [email protected] (this is a more narrowly focused list). Looking ahead, the reliability work is wrapping up (although, much like you, we’re always investing in this area) and you can expect new feature work to start ramping up (for example, we’ll have 64 bit JVM support landing soon). The beta launch of Managed VMs will progress towards General Availability and, in parallel, we’re actively looking at which additional generalized services need to be surfaced into the PaaS layer and how we can make the App Engine experience you all know and love even better. 2015 is going to be a very exciting year! -Dan Sturman VP, Engineering On Sunday, November 9, 2014 9:30:16 PM UTC-8, Vinny P wrote: On Fri, Nov 7, 2014 at 10:17 PM, Brandon Thomson <[email protected]> wrote: Fixing bugs in legacy code is not exciting work and a new generation of engineers at Google may be tempted to "improve" things that aren't broken instead of doing the hard work of maintaining the existing code. +1. New is not necessarily better. To go on a minor tangent, I liked the older Google Groups UI better than the current version. On Sat, Nov 8, 2014 at 2:56 PM, Kaan Soral <[email protected]> wrote: pdknsk has a nice point, you must be on a high support level :) And +1 as well. A paid support contract gives the App Engine unicorns some extra pep in their step :-) ----------------- -Vinny P Technology & Media Consultant Chicago, IL App Engine Code Samples: http://www.learntogoogleit.com -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/google-appengine. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/google-appengine. For more options, visit https://groups.google.com/d/optout.
