Page 1 of 3 123 LastLast
Results 1 to 10 of 25

Thread: Why use sencha and extjs?

  1. #1

    Default Why use sencha and extjs?

    I know this sounds like a troll post but really trying to decided a direction for a small project.

    Looking at Angular and Clarity (VMWare). Vuejs... all the other popular frameworks. We have some people that have used ExtJS in the past and also our parent company still uses it extensively.

    There are some nice things about ExtJs such as it has just about everything we need and would not have to pull in components from other places. The build environment seems to work from my testing. I had one situation where I could not get a scss file to recompile and had to start sencha app watch over. The learning curve seems steep but that is the case with lots of frameworks.

    I guess my biggest issue is just that it is not main stream anymore. Seems like the new company is not going to do much with the product. Finding developers willing to use this tool may be difficult. I had a conversation with a contractor we (parent co) sometimes use and he had the same issue. He is about to start a project with a client that used extjs in the past and they are going with Angular and some other css framwork for a rewrite.

    Any insights would be helpful.

    Thanks.

  2. #2
    Ext JS Premium Member
    Join Date
    Sep 2009
    Location
    Cypress, Kalifornia
    Posts
    306

    Default

    Your positive statements are correct.

    The negative argument is also correct.

    The only challenge I see is integrating multiple components from other vendors/open-source if you go with Angular. You may have a short-term win, but your customer may have future costs. For example an open-source combo-box may not work with future upgrades. Or the other components may have a different look-and-feel. I don't know (one way or the other) if your comment that the new company is not going to do anything with the EXTJS product. In our case we've committed 10 years to it so there's no exit point...except $AP. That may be our direction.

  3. #3

    Default

    Thx for the reply Doug. Our parent is in the same situation. They have their primary software written extjs going back to the open source days. We could use some people over there to help with our project. I do like the platform. Am I wrong to think their MVC & MVVM is well thought out and seems to work? Seems developing for PC/Tablet/Phone would be simplified using ExtJs.

    Do you have any experience with the testing tools and/or sencha architect?

    -troy

  4. #4
    Sencha Premium User mitchellsimoens's Avatar
    Join Date
    Mar 2007
    Location
    Gainesville, FL
    Posts
    40,449

    Default

    If you have a large, or even medium, Ext JS app, I wouldn't recommend moving away from it without a clear plan that you feel very confident you can execute. Ext JS has been great for its fit and finish on its API, component quality and feature set. Even tho there are some buzzword features that are missing like hot module replacement and virtual dom (that are great things), Ext JS can still survive without those and could be made to work with them if enough support behind the needs.

    However, there are some projects that are gaining ground and can even be seen as better than what Ext JS has. The tradeoff, many of them are purpose built components so you have to mix them together and styling can be a huge pain. That doesn't mean you cannot create something fantastic, I'm part of a company that I see this happen daily.

    But you have to also consider the vendor lockin and if you are ok with this. They release progressive roadmaps which sound great but their execution is a concern as their releases thus far haven't had much in them yet (lot of the features were done before old dev team left). I don't see them communicating with many of the questions on the blog or this forum. I understand not talking to me since I've been critical but I've seen customers go without responses which tells me it's not because of me.

    For having a hard time finding devs, that's been an issue for a few years now and has been getting worse and worse. I feel decisions like the GPLv3 license change and 5pk bundle have really hurt this along with still doing things the "old way" and not the new way with new ES specs and tooling around it (e.g. webpack). Their roadmaps seem to be wanting to tackling some of that so only time will tell if they can deliver and what sort of "upgrade" would need to happen.
    Mitchell Simoens @LikelyMitch
    Modus Create, Senior Fullstack Engineer
    ________________
    Modus Create is based on the model of an open source team. We’re a remote, global team of experts in our field. To find out more about the work we do, head over to our website.

    Check out my GitHub:
    https://github.com/mitchellsimoens

  5. #5
    Ext JS Premium Member
    Join Date
    Sep 2009
    Location
    Cypress, Kalifornia
    Posts
    306

    Default

    For having a hard time finding devs, that's been an issue for a few years now and has been getting worse and worse. I feel decisions like the GPLv3 license change and 5pk bundle have really hurt this along with still doing things the "old way" and not the new way with new ES specs and tooling around it (e.g. webpack). Their roadmaps seem to be wanting to tackling some of that so only time will tell if they can deliver and what sort of "upgrade" would need to happen.
    Agreed. But the most vocal people seem to be individual contractors/consultants. The 5K was never an issue for a larger company. Think of the cost of Oracle, SAP, Siebel, etc. EXTJS is a drop in the bucket.

  6. #6
    Sencha Premium User mitchellsimoens's Avatar
    Join Date
    Mar 2007
    Location
    Gainesville, FL
    Posts
    40,449

    Default

    Quote Originally Posted by dougbieber View Post
    Agreed. But the most vocal people seem to be individual contractors/consultants. The 5K was never an issue for a larger company. Think of the cost of Oracle, SAP, Siebel, etc. EXTJS is a drop in the bucket.
    Yep, it really helps to get the solo people to talk about it and learn it so they can look better for the large enterprises.
    Mitchell Simoens @LikelyMitch
    Modus Create, Senior Fullstack Engineer
    ________________
    Modus Create is based on the model of an open source team. We’re a remote, global team of experts in our field. To find out more about the work we do, head over to our website.

    Check out my GitHub:
    https://github.com/mitchellsimoens

  7. #7
    Sencha Premium User
    Join Date
    Nov 2010
    Location
    Chicago
    Posts
    2,425

    Default

    Quote Originally Posted by mitchellsimoens View Post
    ... tooling around it (e.g. webpack).
    I just built a small demo React app using Parcel. I have to say the process is very simple. FYI this developer claims that Webpack is now obsolete.

    https://www.youtube.com/watch?v=hCxvp3_o0gM

  8. #8
    Sencha Premium User
    Join Date
    Nov 2010
    Location
    Chicago
    Posts
    2,425

    Default

    Quote Originally Posted by troy.aumiller View Post
    I guess my biggest issue is just that it is not main stream anymore.
    It's not mainstream anymore because the code base and tooling are getting dated. Ext JS lost a lot of popularity over the last several years and Sencha is not doing much to change this downward trend.

  9. #9
    Sencha User
    Join Date
    Apr 2013
    Posts
    913

    Default

    If you have a big old project made with ExtJS, it might be quite tricky to move to something else, but if you are starting a new one, I personally wouldn't recommend using ExtJS. There are just too many major issues that can not be overlooked at this point:
    • Some time ago ExtJS was split into "classic" and "modern" toolkits. The old classic toolkit supposed to support old browsers (IE8), uses JS for component layouts, which are quite slow, and doesn't work so well on mobile. The new "modern" toolkit is half-finished, missing a lot of features the classic toolkit has and also is quite buggy.
    • Like other people have already mentioned: tooling is ancient and not any good and it can be tricky to integrate third party libraries.
    • When Sencha was acquired by IDERA, they "let go" all the core developers and experienced support people and replaced them by some cheap Indians who have no idea how ExtJS works.
    • Even Don Griffin who was "Senior Director of Engineering" at Sencha doesn't think ExtJS has any future anymore.
    • Extremely long release cycles. Only two stable versions (including minor and major) per year is nothing more than a terrible joke. Not only you have to maintain quite a few overrides to fix all the bugs but the new releases have so many [breaking] changes that it's usually a real headache to upgrade from one version to another.
    • It can take years to fix some bugs [if they get fixed].
    • The paid support is not any good.



    A year ago we have started using DevExtreme Angular for some new projects. If you are looking for a all-in-one GUI framework, I would highly recommend it. High quality support, timely releases, big variety of GUI widgets with pretty good API, good documentation and examples.

  10. #10
    Sencha Premium User mitchellsimoens's Avatar
    Join Date
    Mar 2007
    Location
    Gainesville, FL
    Posts
    40,449

    Default

    Quote Originally Posted by hakimio View Post
    Some time ago ExtJS was split into "classic" and "modern" toolkits.
    Well, more Sencha Touch was merged with Ext JS as the modern toolkit which of course didn't have all the components that classic Ext JS did. There was a lot of work in 6.5.0 to prepare for massive attack in this arena and by now we would have had it next to 100% parity.

    Quote Originally Posted by hakimio View Post
    tooling is ancient and not any good
    Tbh, I've never had any issue with Cmd but maybe that's because I had the luxury of architecting the fullstack to align with Ext JS and Cmd instead of trying to get Ext JS and Cmd to fit into some existing backend. Sencha Touch had node based build tools but this was a stopper for many as, iirc, many IT departments were not allowing node due to licensing. This is why we went with Java for Cmd and now node has become a lot more than what it was back then and you see it being used everywhere now. So ancient... well only because at the time it was one of the few avenues we could go down but I still think it works very well but with everything, your milage may vary depending on what you need/want to do.

    Quote Originally Posted by hakimio View Post
    Extremely long release cycles.
    This was something that bothered me a lot. Yes, we'd have like 500 bugs fixed in each release which is a ton but it took time to fix all of them. There were some talks about doing a scheduled release (like every 6 weeks) which would have been tremendous but it obviously was never more than talks. I believe the stopper was QA (we used Saucelabs for PR testing and outsourced the real QA) but I'm not sure if that was due to the cost of it or if it was due to their testing setup they were creating to automate it also.

    With all the new stuff Idera has on their roadmap, I'd love to see the release cycle reworked too.
    Mitchell Simoens @LikelyMitch
    Modus Create, Senior Fullstack Engineer
    ________________
    Modus Create is based on the model of an open source team. We’re a remote, global team of experts in our field. To find out more about the work we do, head over to our website.

    Check out my GitHub:
    https://github.com/mitchellsimoens

Page 1 of 3 123 LastLast

Similar Threads

  1. Replies: 1
    Last Post: 9 Sep 2016, 5:25 AM
  2. Replies: 3
    Last Post: 6 Jun 2014, 1:49 AM
  3. Replies: 1
    Last Post: 13 Feb 2014, 3:48 PM
  4. Replies: 0
    Last Post: 11 Jul 2012, 5:02 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •