Re: [google-appengine] Re: "One senses GAE is just not a major priority for Google"
Make sure you read this on Wall Street Journal just hot of the press: http://online.wsj.com/articles/google-renews-its-cloud-efforts-1415062792 (Search online for the title "Google Renews Its Cloud Efforts” if you hit the paywall). I quote from the article: "Amazon started renting computing power in 2006, in a unit known as Amazon Web Services. Google unveiled its offering two years later. But Google initially required customers to write software similar to the way Google does. Amazon’s more flexible approach proved more popular. It used so-called virtual machines that let developers use most popular programming languages, databases and other tools. Google adopted this approach later, but by then it had ceded the early lead." "Google’s first cloud service, App Engine, used the company’s internal approach. In contrast, Amazon used more standard technology.” PK http://www.gae123.com On November 3, 2014 at 8:41:15 PM, Robert King (kingrobertk...@gmail.com) wrote: I agree with Jeff Schnitzer. I think google is afraid of the whole open source & docker thing gaining huge momentum and doesn't want to miss out so they want to make that part of their foundation e.g. things like kubernettes. They also want to make it easy for people to migrate from other clouds to google cloud. I personally like app engine instances - they are very lightweight containers - docker is great if you need third party libraries that aren't on app engine & lots of developers don't want to shy away from that. I personally like the simplicity of app engine containers & most of my app can run on them. 95% of my code can run on app engine doing orchestration - I can always have processes running on compute engine with docker etc if I need - but even as it stands app engine can do a huge amount of what you need. Even if google didn't make many improvements to app engine for a while - I can still do a huge amount with app engine. I think a lot of grads coming out of university will go with PAAS & things such as firebase to keep things simple, however a lot of the senior developers will like to stick with what they know. It also seems like appengine needs a faster deployment cycle & better testing / simulation. So perhaps there is technical debt there. On Saturday, 1 November 2014 06:55:34 UTC+13, Jeff Schnitzer wrote: I agree. I thought that article was basically a fluff piece written by someone who has never actually used GAE. Nobody ever cared about the "subset of Java" issue except Sun who, as non-users, count only as whiners ("no, Java's mine, you have to use it the way I want!"). And the very old version of python was fixed (2.7, well, yes, it's still old but let's face it half the Python community hasn't made it to 3.0 yet). IMHO, the biggest issue is that human beings are slow to adopt new things. Most web developers never move beyond the first stack they learn (usually LAMP or Rails). Ask them to go outside of their MySQL comfort zone and they get all nervous and sweaty. GAE is something different, and the truth is that even programmers are a conservative lot. There are real problems with GAE (those two items chief among them) but I think the main reason Google is focusing so much on Compute Engine instead of GAE is that the vast bulk of developers haven't bought into the concept of PaaS yet. They've just barely made the mental transition off of colocated boxes. IaaS is an easier sell, even if it's a dumb choice. Jeff On Fri, Oct 31, 2014 at 5:28 AM, Tapir wrote: On Wednesday, October 29, 2014 5:11:15 AM UTC+8, Emanuele Ziglioli wrote: I would find hard to disagree: IBM, Google, and Oracle are all equally at pains to deliver a message that makes them uniquely attractive. In this regard, Google's inability to recover from the botched roll-out of Google App Engine (GAE) will surely go down as one of the oddest business cases. It launched the product with great fanfare. But developers who flocked to it initially found a difficult platform that supported only a subset of Java and a very old version of Python. Moreover, the interfaces to the proprietary database were poorly thought out, so that almost everything in GAE required platform-specific code-arounds. While GAE has improved in a limited sense since then, Google has not done what Microsoft did — revamp the product from top to bottom to make it easy to use. Nor has it leveraged its natural connection to developers. One senses GAE is just not a major priority for Google. http://www.drdobbs.com/cloud/whose-cloud-will-you-use/240169229 GAE really has two problems, neither of them are belong to what mentioned in this article. On the contrary, what mentioned the article are really good point, IMO. The two problems are: 1. high price, for both instance hours and bigtable operations. 2. long Java instance startup time. In my GAE experience, it is very reliable. BigTable is very powerful and easy to use. -- You receive
Re: [google-appengine] Re: "One senses GAE is just not a major priority for Google"
I agree with Jeff Schnitzer. I think google is afraid of the whole open source & docker thing gaining huge momentum and doesn't want to miss out so they want to make that part of their foundation e.g. things like kubernettes. They also want to make it easy for people to migrate from other clouds to google cloud. I personally like app engine instances - they are very lightweight containers - docker is great if you need third party libraries that aren't on app engine & lots of developers don't want to shy away from that. I personally like the simplicity of app engine containers & most of my app can run on them. 95% of my code can run on app engine doing orchestration - I can always have processes running on compute engine with docker etc if I need - but even as it stands app engine can do a huge amount of what you need. Even if google didn't make many improvements to app engine for a while - I can still do a huge amount with app engine. I think a lot of grads coming out of university will go with PAAS & things such as firebase to keep things simple, however a lot of the senior developers will like to stick with what they know. It also seems like appengine needs a faster deployment cycle & better testing / simulation. So perhaps there is technical debt there. On Saturday, 1 November 2014 06:55:34 UTC+13, Jeff Schnitzer wrote: > > I agree. I thought that article was basically a fluff piece written by > someone who has never actually used GAE. > > Nobody ever cared about the "subset of Java" issue except Sun who, as > non-users, count only as whiners ("no, Java's mine, you have to use it the > way I want!"). And the very old version of python was fixed (2.7, well, > yes, it's still old but let's face it half the Python community hasn't made > it to 3.0 yet). > > IMHO, the biggest issue is that human beings are slow to adopt new things. > Most web developers never move beyond the first stack they learn (usually > LAMP or Rails). Ask them to go outside of their MySQL comfort zone and they > get all nervous and sweaty. GAE is something different, and the truth is > that even programmers are a conservative lot. > > There are real problems with GAE (those two items chief among them) but I > think the main reason Google is focusing so much on Compute Engine instead > of GAE is that the vast bulk of developers haven't bought into the concept > of PaaS yet. They've just barely made the mental transition off of > colocated boxes. IaaS is an easier sell, even if it's a dumb choice. > > Jeff > > On Fri, Oct 31, 2014 at 5:28 AM, Tapir > > wrote: > >> >> >> On Wednesday, October 29, 2014 5:11:15 AM UTC+8, Emanuele Ziglioli wrote: >> >>> I would find hard to disagree: >>> >>> *IBM, Google, and Oracle are all equally at pains to deliver a message that makes them uniquely attractive. In this regard, Google's inability to recover from the botched roll-out of Google App Engine (GAE) will surely go down as one of the oddest business cases. It launched the product with great fanfare. But developers who flocked to it initially found a difficult platform that supported only a subset of Java and a very old version of Python. Moreover, the interfaces to the proprietary database were poorly thought out, so that almost everything in GAE required platform-specific code-arounds. While GAE has improved in a limited sense since then, Google has not done what Microsoft did — revamp the product from top to bottom to make it easy to use. Nor has it leveraged its natural connection to developers. One senses GAE is just not a major priority for Google.* >>> >>> >>> http://www.drdobbs.com/cloud/whose-cloud-will-you-use/240169229 >>> >> >> GAE really has two problems, neither of them are belong to what mentioned >> in this article. On the contrary, what mentioned the article are really >> good point, IMO. >> >> The two problems are: >> 1. high price, for both instance hours and bigtable operations. >> 2. long Java instance startup time. >> >> In my GAE experience, it is very reliable. BigTable is very powerful and >> easy to use. >> >> >> -- >> 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 google-appengi...@googlegroups.com . >> To post to this group, send email to google-a...@googlegroups.com >> . >> 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 google-appengine+unsubscr...@googlegroups.com. To post to this group, send email to google-appengine@googlegroups.com. Visit this group at http://groups.google.com