Why is Webix so little known?

edited December 2014 in General

Compared to other UI widget libraries, Webix is very, very little known, despite the widgets being really powerful, and despite a very compelling demo at http://webix.com. So why is that? Why does Webix have only 12 stars on GitHub and only 9 questions on StackOverflow, two of which are mine? Why are there no articles in "reliable sources", as Wikipedia calls them, about Webix, so that the Wikipedia article is in danger of being deleted as not notable enough?

Here are some ideas to improve this situation:

  1. Spell-check the entire docs.webix.com site, and fix all the dead links. The situation is pretty embarrassing now - during the past workday, I've filed on GitHub over 10 issues regarding documentation errors.
  2. Once that's done, hire an editor fluent in English and software development (sites like proz.com are great to find editors), and have that person go through the documentation and make it professional. Documentation matters. Kendo has excellent documentation.
  3. This forum is nice, but it's a very isolated community. Direct instead users to ask Webix questions on StackOverflow, a place that will expose Webix to a lot more front-end developers. Keep the forum only for general discussions that would be too unspecific for StackOverflow.
  4. Answer questions on StackOverflow, Reddit, Quora etc. about UI widget libraries, and point to Webix, explaining why, along with details (I've seen Quora answers promoting Webix, and some of those were rather empty promotional answers). It's sad that people still wonder "Is Twitter's Bootstrap the only thought out/complete front end UI library?". But Bootstrap got marketing right.
  5. Help me integrate Webix with Meteor.js. Meteor is the most popular full-stack JavaScript framework on GitHub (over 21,000 stars), but it doesn't have a widgets set. Integrating Webix with Meteor will bring a lot of exposure to the library. (By the way, Meteor.js directs their users to ask questions on StackOverflow, which has helped the popularity of the framework.)

I'm writing this because I work at Google and I'm evaluating Webix for use in an open-source project sponsored by Google. I would love to be able to recommend Webix whole-heartedly.

Comments

  • edited December 2014

    I wouldn't disagree with anything you say, from what I can see Webix is only a small company, so resources are finite. But a couple of suggestions I would add

    1) Give more examples, especially related to integrated CRUD controls for Datatables and trees, along with backend scripts.

    2) Webix has built a wealth of example on the CodeSnippet, make this searchable and build an organised KB from it.

    3) There is a wealth of Forum posts, many of them asking the same question, this is searchable, but likewise add these to an organised KB.

  • edited December 2014

    Hello, Dan. What concerns stackoverflow, you've found only the questions tagged with [webix]. But not all people tag their questions. So here you can find other questions about the Webix library: http://stackoverflow.com/search?q=webix. Be more attentive, please, before writing such a post.

  • edited December 2014

    Oh wow, 25 instead of 9. I'm floored.

    No matter how attentive I am, that doesn't even compare with, say, Meteor.js, which has between 8,000 and 19,700 results on SO, depending on how you count. Meteor was launched in April 2012.

    Nika, might I respectfully suggest hiring a native English speaking technology evangelist for Webix? Someone like John Lindquist for Jetbrains (he does a great job presenting WebStorm, while the product is developed in Russia). Webix is a great library, I've drastically updated its Wikipedia page to reflect that (you're welcome), but it needs someone to represent it in the open source community with a style and tone that appeals to a large audience. I come from Romania myself and patterns like "Be more attentive, please, before writing such a post." don't bother me that much; however, a US-based CTO evaluating Webix could be put off by a statement of that sort. Defensiveness doesn't sound good. In this case, kelv1n and I would be happier to hear that our suggestions will be taken into account, and truly happy if something was done - even something minor but significant, such as addressing some of the trivial issues I've filed on GitHub over the weekend (e.g. typos or broken links in the docs).

    Just my two cents. For example, Maksim's answer to my tree icons question was great, and kept me from switching to Kendo.

  • edited December 2014

    dandv, again I don't disagree. But Webix is in a difficult position and competing against long established alternative projects, so it takes time to build. Likewise hiring somebody can be very costly, especially an experienced technical writer who has a solid background in development.

    Nika, I think Dandv has a valid point aswell, take community suggestions for what they are, just suggestions :) then give feedback about what you guys consider a good idea and will look at adding.

    Another couple of idea

    1) Give a road map

    2) Provide a "sneak peak" of future developements

    3) Give some more examples of integrating with requirejs and dhtmlx (Maksims example of requirejs was amazing, way beyond my abilities and i'm trying to decipher it)

  • Hello Dan and kelv1n,

    Thank you for your suggesstions. I'm glad that more and more people get interested in Webix.

    Your ideas are really helpful and effective. We will consider them and probably implement some of them in future.

    Dan, thank you for your ideas in promotion of Webix and active participation in forum life.

    kelv1n, we'll consider your idea about giving more examples of integrating with requirejs and dhtmlx and add a separate article on this topic in Webix blog.

    We are constanly improving the library code and documentation, thank you so much for reporting bugs and typos.

  • edited December 2014

    Thanks Nika, I only just looked at the blog for the first time lastnight. I land on the front page and go directly to the Forum or API documentation.. sorry couple more suggestions

    1) Add a "NEWS" for blog posts etc box to the top of your landing page

    2) Documentation is a pain to maintain, have a "report correction" button on each page that users can simply click and enter the correction

    3) Setup a community WIKI

  • edited December 2014

    I was going to suggest the same thing re. the documentation - offer ways to contribute. If you put the documentation on GitHub, I volunteer to correct many of the typos.

    Maybe use the gh-pages branch and serve it under the docs.webix.com subdomain, or fetch it from GitHub and publish on your own; but the idea is to make the documentation easy to contribute to.

    Here's how MongoDB does it - http://docs.mongodb.org/manual/. You see an edit icon in the top right, and clicking it takes you to GitHub where you can edit that page of the documentation.

  • edited December 2014

    Nika

    Another suggestion - Would it be possible to update the skin builder to save themes, so they can be shared through the webix site (and build a gallery from them).

    Webix has some amazing tools available, by simply adding a few community aspects to them and allowing people to share, it will greatly enhance how they get used, which will encourage more users to get involved, and you'll find the userbase which grow organically :)

  • In my eyes, it became more difficult to be under the spotlights as with mentions from others :

    • IT Inertia
    • market already a bit saturated
    • competing against long established alternative projects
    • Short of breath developers to have learnt the framework in the buzz of moment, then that must fall over towards the next

    framework into the buzz before even the end of their learning of the first one and so on, after a while, they leave grip of the frenzy of this forward perpetual and too fast push.

    There's also the fact than many UI widgets libraries are really close competitors, ie they do more or less the same things and share quite the same goals. They want to get traction but do they share what the audience want from them ?

    Each time i see a new candidate, i can closely compare it to another one, x or y already on the market for quite some time. Each time, i hope it will take a different route but instead, i always see it taking the same evolution path as one of the existing ones. If you go into the crowd, who will see you ? Will you see your audience from there ? And for whom will you be useful ? the crowd itself or the audience ?

    To be noticed, imho you have to bring something significantly different which answer real needs to the people you target as your audience and for which almost no one did something for.

    The audience is generally composed of developers and they are not stark naked, ie they already made some technical choices, you have to deal with this background. I'm talking about integration here and this is neglected most of the time. The audience is people tied to a framework culture; they choose for example a server side framework, they would like to adopt a
    given UI widget library but they have nothing durably strong to link the two with efficiency and will have headaches and will take holidays in trying to make this work.

    Who would say no to product bringing this integration so the audience can focus on his real work ?

    Second point is to make things more easy. Who would say no to product proposing to achieve things more easily so they would have a more happy developer life, more confort, and allow them to react faster against their goals ?

    Instead of this happening, we always see yet another product without any of this. So instead of developping yet another new product, maybe more interesting to build the integration link between two existing products which need and benefit from this.

    That's why i think, in order to be seen and followed, a beginning of answer tied into the equation "ease * [studio + integration]".

    Ease, confort and pleasant can begin with a builder such as sencha architect. There have been too few examples in this area. And for the talk, can we say sencha architect, vaadin, smartclient and a few others are a failure in attracting people ? I don't think so.

    Regarding integration, i did'nt see often something done on the subject. ExtJS4Yii linking Extjs with Yii php framework is one of the very few examples. I saw more trying to achieve the same thing (without success) and it always get a lot of traction but the ones who tried were never the authors of the product for which the integration has to be built.

    At the end, everyone found his wonderland in each area and can't make them work together. The list will be too long so i will take recent examples :

    Where is our UI widgets library and DB supported by products such as meteor or derby ? or a php framework, or a widely adopted cms such as wordpress, clearly not bad areas where to get traction from.

    Why no UI widgets library authors dive into Wakanda integration ? while wakanda have a studio they opened in such a way almost everything possible has been done for making external UI widgets library to be integrated into it. Or no one either participating into an effort to make the studio part of wakanda to be opened to things different to wakanda server, such as meteor for example. Wakanda is not against this, they even propose their support for doing so.

    I don't know, i believe UI widgets libraries authors and db are not ready to take "marketing" opportunities and focus more on development itself.

    I don't know why core developers of new products don't bring what people are waiting for, but surely they have the good recipe to not get traction, and this is a so old recipe to say the least.

  • I think Webix integrates nicely with many players in the ecosystem - check https://en.wikipedia.org/wiki/Webix#Integrations

    Also, learning it was pretty easy for me, compared to Dojo (from what I've heard - many have mentioned Dojo's and ExtJS's steep learning curve).

    It also does some things well that ExtJS refuse to do - e.g. having both the code and the documentation on GitHub, and accepting pull requests.

    Where is our UI widgets library and DB supported by products such as meteor or derby ?

    http://forum.webix.com/discussion/3715/advice-for-integration-with-meteor

  • @dandv - I've been using Webix for about a year, and have also been wondering why it's not better known - I haven't found a UI toolkit that has anywhere near the functionality, performance, and browser support.

    If IE8 is a requirement (yes, I know you guys at Google would love to deprecate that browser, but it is still prevalent in corporate US), then Webix is an easy choice.

    I also agree that the documentation might have put off several potential users, it needs some serious work, specifically:

    a) a native English technical writer needs to correct the numerous spelling and grammatical errors,

    b) the 'flow' through the documentation is not clear for beginners, too many potential starting points, with overlapping information, that needs to be fixed.

    c) while the demos, api, styling info, etc. are mostly extant, if you look hard enough, the documentation really needs a decent framework. Your suggestion of using the Mongo documentation as a template seems reasonable.

    I have also thought about contributing to the documentation. If Maxim puts up a documentation site, using some well established template (Mongo, or other), I'd be willing to work on it too.

    Given that you and I seem to be based in the Bay Area, I wonder how Maxim would feel about us starting a Webix Meetup. I'm in Sausalito, presumably you're on the Peninsula, so SF would be the place to do it. There are Meetups for many lesser frameworks, no?

    David

    ps: the other thing Webix needs is MVVM support, like GluJS created for ExtJs.
    That developer is local btw (to the Bay Area), I tried to hook him and Maxim up a few months ago, without much success I have to admit. The Firebase integration might be the spark that gets MVVM implemented in Webix.

  • Hey @dmoshal - in the meantime Webix has put their docs on GitHub at my request - https://github.com/webix-hub/docs. I've submitted a bunch of improvements, mostly having to do with dynamic data loading, and went through the first-time contributor pain (e.g. Markdown peculiarities, or what files not to edit). Contributing should be quite straightforward now.

    a) - Yes. @Nika, I would suggest looking to hire on contract an English IT editor on ProZ.com, unless you can mount enough open source interest so people come and fix the documentation for free.

    b) - Yes! There should be one starting document that explains how a component is built up. I.e. the common properties such as "type" and "css" that apply to all components, and how Webix interprets them.

    I think Maksim would be more than happy if we started a Meetup. I'm in the South Bay and I've drummed up interest at the Meteor SF meetup with this Meteor-Webix ANNOUNCED integration, which (haha) has gained in one week more GitHub interest than Webix itself. #marketing

    MVVM might come from the Meteor integration direction.

  • @dandv, that's excellent news, I'll take a look at the doc.

    Before we get too far into editing those files it probably makes sense to agree on a documentation framework (with Maxims approval). One that has a decent flow, and an efficient way to navigate the api and styling attributes.

    I'd suggest that we propose a couple of other open source projects as potential candidates.

    Re: the native English editing.
    Once we have the framework, editing the spelling and grammar should be straight forward. I'm willing to contribute there.

    Re: the Meetup.
    South Bay could work, SF would be better for me.

    Re: Meteor.
    I've been building, and delivering, reactive web-based apps since 2000 (yes, it was possible, using hidden iFrames, long-polling, followed by the then secret HttpRequest, etc.), so I have some idea of the real-world problems lurking in that domain. In my view there are fundamental problems with their approach, I'll just leave it at that.

    On the other hand, Firebase, which was acquired by Google last year, has been far more successful, by focusing on very specific parts of the reactive web problem. There is already an early Firebase / Webix integration with basic functionality.

    My suggestion, for what it's worth, is to integrate Firebase with Webix using an approach similar to GluJs, with declarative data and attribute binding, reactive functions, etc.

    I actually discussed this concept with Michael Gai, the GluJS developer (who really understands the issues in this domain), six months ago, and he seems open to the idea, see my exchange with him last year:

    https://groups.google.com/forum/#!topic/glujs/qQjW_ByL5LU

    anyway that's by 2cents.

    Dave

  • @dmoshal: Webix docs already use a framework, apparently a custom-built one called "madoka".

    Regarding the Meteor integration, we might have some licensing problems.

  • @dandv: another reason to integrate with Firebase instead.

  • well, with thanks to Dan, Webix is not that 'little known' anymore. :)

    Have used it(webix) for some personal apps and it works more like Okay for me.

    UIs are getting ready faster than ever. So i would definitely vote for more Meteor integration. Then i'm buying it(webix).

  • edited August 2019

    Hey Webix team, any response to this, five years later?

    Let's focus on one thing, the docs. I'm still repeatedly running into annoying problems with the documentation.

Sign In or Register to comment.