Anybody else want to commission a open source Wordpress type tester?

24

Comments

  • I'm interested in helping develop a module like this as an open source effort
  • Interested . . .
  • I have to admit that my mind is a bit blurred at the moment because of a very nice Rioja from 2011, but if I understand it all correctly (probably not) then I can participate and put money into the development of this open-source Type Tester, so that I can use it, or I don’t spend money and then can use it still because you are paying for it?
  • If you pay for it, then it will support the features you need, or if you don't pay for it then it might not.
  • But if I pay for the tool and hence it supports the features I need, why would I give it away for free then (sorry, much too much Rioja now)?
  • Erin McLaughlinErin McLaughlin Posts: 45
    edited July 2016
    Sure! Keep me in the loop
  • Okay, I’m in too (I’m drunk anyway).
  • James PuckettJames Puckett Posts: 1,967
    Ok, so what features would we need?
  • I reconfirm my interest.
  • Johannes NeumeierJohannes Neumeier Posts: 356
    edited July 2016
    Feel free to add or edit this list, but to get things rolling... off the top of my head:

    Interface options:
    * single line vs paragraph previews

    * input = editor vs linked separate input field that manipulates preview
    * font size switcher
    * line-height switcher (?)
    * letter-spacing switcher (?)
    * customisable style and layout
    * dynamic font loading (and switching)
    * shorthands for changing the text sample (default options, testing strings)
    * activate different opentype features (browser support?)
    * provide image fallbacks
    * support these webfont formats: woff, woff2, (?)

    Backend / plugin options:
    * webfont obfuscation / character subsets (?)
    * wordpress plugin with font file upload, manager and page integration shortcut
  • Dave CrosslandDave Crossland Posts: 1,389
    @Behdad Esfahbod might have something helpful to say about the "provide image fallbacks" aspect; the 'hb-view' program in the harfbuzz package provides PNG rendering with full OT processing. 

  • Lars SchwarzLars Schwarz Posts: 114
    Dave: hb-view probably won't run on most shared hosting environments, so if you want to stay indie and not use a service that hosts hb-view to serve image previews that could only be make available if the server that WP install runs on has hb-view installed.
  • Lars SchwarzLars Schwarz Posts: 114
    Happy to join the dev side 
  • Dave CrosslandDave Crossland Posts: 1,389
    Dave: hb-view probably won't run on most shared hosting environments
    This seems a little antiquated to me, in the age of container hosting environments. 
  • Johannes’s list looks very comprehensive from the user point-of-view. Maybe too much so? I could dispense with being able to customize the layout. How far would it go? Changing the alignment? Changing the margins, the number of columns? 

    With so many options it might be difficult to keep the user interface clean and attractive. I'm not a fan of default hidden controls as on the Klim site (I like to know up front what my options are), but others might prefer them so default showing/hiding of controls should be a setting option on the back end.

  • Personally I agree that simplicity should be paramount. My list above rather serves to keep track of possible features, so not everything is needed or useful enough to warrant the effort of building that feature. Also note that different weights / styles is in effect covered by dynamically loading and displaying different fonts, even though it might be a separated issue in the UI.

    Splitting this into smaller parts, the effort could proceed like this:

    * Build a javascript module with generic functionality, for example a jQuery plugin, that handles the functionality, but doesn't specify the interface implementation or look (people can use this also for other projects or platforms then)
    * Make a sample interface using that module and possible other UI modules, with a generic style and usability in mind
    * Make a Wordpress plugin that uses this interface to render webfonts added via the Wordpress backend, e.g. by defining and adding shortcodes to display a particular tester with particular defaults and options
  • Lars SchwarzLars Schwarz Posts: 114
    Dave: hb-view probably won't run on most shared hosting environments
    This seems a little antiquated to me, in the age of container hosting environments. 
    Docker for a simple WP instance and ssh into it to add hb-view? Well, yeah, if your foundry site requires such scalability, but if you use WP for your foundry site you probably don't need clusters and stuff. Just saying that from my point of view you should build something that works out of the box with 99% of all WP installs.
  • Not sure how much time I really can commit, but to get the ball rolling I started a repo for a jQuery plugin and started some simple manipulation and loading code for type previews.

    Feel free to hack away at it ;)
  • James PuckettJames Puckett Posts: 1,967
    I have my doubts how good any tool that tries to cover everyone’s particular needs can become. 

    That’s the point of making it open source. Have one tool with basic functionality so people aren’t doing that work again and again. Then people with particular needs only need to pay to address those needs.

  • @Frode Bo Helland Can you elaborate on both the points you mentioned? Do you mean language tags in the sense of the tool would allow the admin to set (or even detect) what languages a specific font file supports?
    What's the difference between inline and global OpenType features?
  • > in the sense of the tool would allow the admin to set (or even detect) what languages a specific font file supports?

    I think a robust solution for this is highly unlikely to be covered by such a project. Language data is spread over multiple sources, and no single initative I’ve found so far does a good job of collecting even a small subset of written languages.
    Right, but is that nonetheless what you meant with language tags, i.e. that there is some option to show the user testing the font what languages are supported - or was this connected to testing language specific OpenType features you mentioned, so that the user has the option to select which language she'd be typing?
  • Hi Roel! As you can see we're exploring the concept which means your client in this case is a committee. I may suggest you start a more private discussion thread solely for those who are committed financially to participating instead of trying to distill feedback from those who aren't.

    Beyond that, it would be great to learn more about your organization and similar projects you've done to what we're discussion.
    I was a bit excited to see the basics being formed of a project that tickles my fancy on many levels, so I might've jumped in a bit too soon.

    But in short, I'm a front-end developer and consultant (some stuff here and here) working on websites and apps (small and large). Accessibility, performance, cross-browser robustness is what I do, type and font tech is what I love to do more. I'll pitch myself more fully when it's more on topic! 
  • Not sure how much time I really can commit, but to get the ball rolling I started a repo for a jQuery plugin and started some simple manipulation and loading code for type previews.

    Feel free to hack away at it ;)
    Would it be possible to avoid jQuery and getting rid of a dependency right away?

    I guess the answer to that would be directly related to "Which browsers should be supported?" Development will look entirely different if the requirement is "IE6+" or "Only the last two versions of major browsers" :)
  • For a font developer it is important to check whether all OpenType Layout features work properly, i.e., according to the specs, irrespective of whether or not applications support these features. For instance OTM can be used for extensively checking this. However, for the end user or the graphic designer who develops corporate identities it is more important to know which features are specifically supported by which applications. After all, if one can’t use certain features in certain environments, there is clearly no need for them. So, what I would like to see implemented is a pull-down menu that shows a listing of (different versions of) mainstream applications (especially Adobe’s suite and Office), which results in the applicable activation of supported features.
  • Would it be possible to avoid jQuery and getting rid of a dependency right away?
    I simply started off with what a) I somewhat know and b) most wordpress sites use anyway. Feel free to lay out the foundation for something non-jqueryish and I'll gladly help work on that instead ;)
  • Dave CrosslandDave Crossland Posts: 1,389
    People using some blogging wp host aren't this tools audience in my opinion. A dockerfile that describes how to install a httpd, php, mysql, wp, and hbview seems straightforward to me, as is running it on gcp, aws, azure, digitalocean, etc etc.
  • Since this regards only the "image fallback" option that some already regard as non-pertinent it's maybe not the most essential of questions if or if not support for this is implemented right away, and how this would affect a possible user base. For example, it could be an option that is only available in the Wordpress panel if those libraries are indeed installed on the system.

    Is there anyone here following this discussion that has experience with authoring Wordpress plugins?
  • Lars SchwarzLars Schwarz Posts: 114
    From my point of view uploading the fonts and/or generating web font kits via that plugin should be discussed in advance, because you could spent much time on adding this functionality to the plugin (or not). I presume most people here know how to generate web font kits and prefer to have as much control as possible when generating them, so there's probably no need to have that functionality in the plugin itself. 
  • @Frode Bo Helland Could you provide me with a woff file that has such features and tell me what to test with, e.g. "try this and that with this and that language, and the difference in rendering should be this"...? Or if you know a Google Webfont or Typekit font that does have such behaviour I could used to implement and verify a feature like that in the javascript implementation?

    @Lars Schwarz With generating subsets or such I was mostly referring to generating webfont files that don't necessarily contain the entire font (both for DRM and loading performance reasons). Not sure if this is an actual thing or more like just my assumption. I agree with you that the target users of this plugin would probably proficient enough or keen on creating the font files themselves. So either way this might not be a less important feature.
Sign In or Register to comment.