Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Ben Blom


Ben Blom
Last Active
  • Re: Plex; IBM's new font identity model

    Making free fonts, facilitates to kill the market of paid fonts.

  • Re: Plex; IBM's new font identity model

    Why did type designers lose?

    It’s the same as with Google Fonts and other big free font schemes.

    The bigger (in relative terms) the market of good quality free fonts becomes, the smaller (in relative terms) the market of good quality paid fonts becomes. It may not be a zero sum game—but the more good quality fonts become available for free, the less people are willing to pay for good quality fonts.

    So type designers who are dependent for their income on the market of good quality paid fonts, will loose out by the increased availability of good quality free fonts.

    The only way to prevent this to happen, is for type designers to show solidarity with their fellow type designers—i.e. not to collaborate with the creation of good quality free fonts. Therefore, it is in the long-term interest of all serious type designers, not to collaborate with any big free font scheme. So, fellow type designers: please don’t shit in your own nest.

    So indeed:
    To those who collaborated [by creating good quality free fonts]: mind the costs of making money.

    The short-term profit of a few, may cause the long-term loss of many.

  • Re: EULAs: No Modifications Clauses.

    Hrant H. Papazian: It’s high time to start concertedly shaming EULAs with a no-mod clause.

    Is this a call for intolerance—by someone who advocates diversity and tolerance? The contents of EULAs are, in the end, a business decision. It doesn’t make sense to draw the shame card for this.

  • Re: Dutch IJ with dots

    it would be worthwhile for the Dutch language community to make the effort to get used to it

    It is silly for someone from outside the Dutch language community, to tell the Dutch how to improve their spelling.

  • Re: I trained a neural network to kern a font (mostly)

    Ben: I’m not sure why you say that this ignores the spacing. The spacing data is there in the sidebearing contours

    Simon, I’m sorry. I didn’t read careful enough.

    If it’s 99% accurate over 160,000 kern pairs that’s still 1,600 mistakes.

    The question is: How bad are those mistakes? If those mistakes are nothing more than a slight deviation from what the kerning should be, then those mistakes are not a real problem. (Perhaps many of those mistakes only concern uncommon pairs.)