Google Fonts: Your Questions, Answered

1911131415

Comments

  • Ben BlomBen Blom Posts: 250
    Vernon: So do you think your revenue from usage of your webfonts will be higher in 2016 than it was in 2010? or lower? or the same?
    I have no idea. Please see my comment of December 5 (Page 4).
  • Dave wrote: "That's why its about LIBRE fonts. The gratis aspect isn't interesting to me, at all, and I find it tiresome when people focus only on this and overtly say that the freedom in libre fonts is worthless."

    Certainly, the value in libre fonts is in the freedom. But then again, the perceived threat is in the price. Hence the tiresome focus on price by those who feel threatened (or like me, who are not personally threatened but see a likelihood of damage to the font industry).

    Vern: As we are only around 25% adoption of real web fonts, I expect everybody's sales are growing. I actually agree that it is almost impossible to be absolutely certain what effect Google's entry is having on the sales of commercial fonts (and it might vary by foundry).

    I'm pretty sure that Google's work is accelerating total adoption rate of web fonts. The real question is, is Google's market share (+ self-hosting of libre fonts sponsored by Google) only proportional to the added usage? Or is it more, or less? Google might still be taking market share that would have gone with commercial approaches.

    Moreover, once we are at the point where adoption would have been at or close to maximum even in the absence of Google, then Google owning any portion of the market that would otherwise have gone commercial is "bad" for the rest of the market.

    I gather Dave believes that Google playing in this space will also cause some folks to try Google's web fonts who wouldn't otherwise have tried web fonts at all, AND that some of those will eventually graduate to "paid" fonts.

    For Google to have a net positive impact on the commercial font industry, not only does Dave need to be right about that, but the size of the effect needs to be greater than the value of market Google is capturing that would otherwise have gone with retail options. This seems pretty unlikely to me, and especially so once we have a mature market (in five or ten years).
  • Thomas; i think there's a good argument that without the google fonts start, the adoption of webfonts as a whole would have been much slower. The evidence seems to show that the google libre library was a significant catalyst for the current levels of non-libre webfonts.

    You seem to paint a picture where non-Libre webfonts were already established on the webfont stage, and then the google library came along, and maybe threatened / or did not threaten sales of those non-libre webfonts. That's a misnomer; it wrongly assumes that the non-libre webfont market was already well established before libre fonts entered the stage.
    I think the truer picture is that non-libre fonts are attempting to capitalize on a potential market that libre fonts created first.
  • Thomas, you overlook the free tier of the proprietary web fonts services. I believe most of the gf traffic comes from blogs and very low traffic sites which would otherwise be costing you money on your loss leader tier.

    http://en.m.wikipedia.org/wiki/Loss_leader
  • Dave, for a loss leader to be effective--or to be considered a loss leader at all--there needs to be a second- and third-stage marketing plan. In other words, a means by which the marketer can follow up on the the giveaway or discount: "You got the starter font for free, now click here to see the better ones (for which you have to pay)." If Google made available email lists of all downloaders of their free fonts to the people who made them, then one could see participating in their program as an extraordinary business opportunity based on a loss leader. But without that, one could reasonably argue that the commercial font firms that participate in the giveaways will retain an advantage, because the large font vendors who give away some fonts, such as Adobe or Monotype, are perfectly capable of helping themselves to the data they need to make future sales contacts.

    Some on this thread who have give away fonts see their doing so as a kind of religious or political act, but for those for whom it something less utopian, I hope they see how their chance of helping themselves under the current structure is quite slim, whereas their chance of helping big corporations is considerably greater.

    It wouldn't take much capital to structure this better, to create a kind of free-font union (say, for type designers under the age of 30) that gathers marketing data for its participants. I bet it would get good publicity, too. You could still do it. But as it's structured now with Google, the poor are destined to stay poor.
  • vernon adamsvernon adams Posts: 82
    edited December 2013
    Certainly, the value in libre fonts is in the freedom. But then again, the perceived threat is in the price. Hence the tiresome focus on price by those who... like me... see a likelihood of damage to the font industry
    Thomas,
    do you mean that Libre fonts in general could do "damage to the 'font industry'"? or specifically that Libre webfonts could do "damage to the font industry"?

    Also, might there be 'Libre' other things that could therefore also do damage to the font industry? such as Libre font engineering tools? or Libre font creation software? or Libre font server software?, or Libre font rendering software?

    Also, how do you define the 'font industry'. It's a term that we hear and use a lot, but i'm not sure what it actually encompasses. For example, when is something 'part of' the 'font industry' and when is something 'outside' of the font industry? Are Libre webfonts outside the font industry? or part of the font industry?
  • John HudsonJohn Hudson Posts: 2,946
    edited December 2013
    A designer either wishes to make a libre font, or they don't, so i think your idea of some kind of 'contractual obligation' to release under a libre license is misleading wording.
    Vernon, both you and Andres seem to be referring to open licensing of fonts in general. I was referring specifically to fonts whose libre release is funded by Google Fonts under contract; I was referring specifically to the wording of the Google Fonts standard contract, which I have read. It is an agreement that, in exchange for money, the font maker will release the font under either OFL or Apache 2.0 license; it is a contractual obligation of that agreement to do so. Failure to do so would be in breach of contract.
    The owner of a libre font can develop and release proprietary versions of their libre font, aka release versions of their OFL'd fonts as fonts not published under the OFL. Obvious, i would have thought.
    That isn't at all obvious from the Google Fonts contract, which states that
    Contractor will release all font files, associated documentation, and other materials created, conceived, prepared, made, discovered or produced by Contractor as part of the Services or any derivatives thereof (the “Deliverables”) under an Accepted Open Source License.
    That suggests to me that there is at least a legal ambiguity in the right of the font maker to release derivatives of the specified fonts under a non-open license.
  • vernon adamsvernon adams Posts: 82
    edited December 2013
    That isn't at all obvious from the Google Fonts contract, which states that
    Contractor will release all font files, blah blah blah. That suggests to me that there is at least a legal ambiguity in the right of the font maker to release derivatives of the specified fonts under a non-open license.
    Don't sign it then! ;p
  • The OFL 1.1 FAQ doesn't seem to address commercial "pro" versions, although that seems at face value to go directly against the spirit and probably the license wording itself. Would be interesting to get opinions on this. Subject for another thread?
  • John HudsonJohn Hudson Posts: 2,946
    edited December 2013
    It seems fairly clear that if one has already developed a font, then released some part of it under OFL, then one remains free to produce derivatives from the pre-existing data that could be released under any form of license. This is the sort of situation that Andres comment seems to address.

    I'm specifically wondering about work produced under conditions defined as Services in the Google Fonts agreement, which would have an impact on whether it makes more sense for a given font maker entering into that agreement to choose the OFL or Apache 2.0 license. As I said, it seems ambiguous, and probably something that should be clarified in writing by Google (which is why I added it to Dave's question list).
  • re: the OFL. You seem to be confusing the license holder and the license user. The license holder of an OFL'd font can release that font, however they want, and under whatever other license they want. The license user though must release derivatives of an OFL'd font under the OFL.
    re: contracts between private parties. I don't see what headway you are aiming at with trying to open up those matters here.
  • John HudsonJohn Hudson Posts: 2,946
    BTW, even if the folks at Google Fonts interpret their contract as meaning font makers producing and releasing works as Services for Google retain the right to release derivatives under non-open licenses, there's no guarantee that Google's lawyers will interpret it in that way.
  • I don't see what headway you are aiming at with trying to open up those matters here.
    Here is specifically a thread about Google Fonts, in which Dave solicited questions for Google Fonts that folks would like answered. It seems to me entirely relevant to ask what the implications of the conjunction of the Google Fonts contract and OFL license are...
    i'm not asking this to be difficult:
    Yes i understand the mechanics of your question, i was questioning your aim, as it's not clear;
    a) Are you sat scratching your head, unsure whether to sign such a contract?
    b) Are you wanting to alert others to something you think they should be aware of if faced with such a contract?
    c) trying yet another angle of attack on the google fonts?

    ps - i'm not asking to be difficult either :)
  • "ps - i'm not asking to be difficult either :)"

    ...despite the hundred rhetorical questions this week.




  • "ps - i'm not asking to be difficult either :)"
    ...despite the hundred rhetorical questions this week.
    hahah :) good one.
    Well... Mr JH did ask me very early on to ask questions of him if i felt i couldn't follow his logic, so i do feel obliged to ask.
  • Nick ShinnNick Shinn Posts: 2,129
    Vernon, perhaps John is just interested in knowing the nuances of how the situation stands, rather than trying to best someone else in an argument over ethics. I find it hard to take sides, having been down this road before with the issue of bundling of free fonts by mega corporations.
  • John HudsonJohn Hudson Posts: 2,946
    Vernon, mostly (b). I'm looking at the Google Fonts contract and trying to figure out whether, if a font maker does enter into this agreement, what options then exist for alternative monetisation of derivative fonts, considering, for example, the possibility that Google won't fund character set extension, additional weights, etc.. One option would be for the font maker to go looking for another patron to fund further development under an open license. But I'm trying to figure out whether the Google Fonts agreement would also permit sale of non-open licenses to such extensions. The agreement isn't explicit in this regard, and the wording regarding deliverables and derivatives seems like it could be interpreted as prohibiting this.

    Definitely not (c). I'm not attacking Google Fonts, and don't think I have done so. I don't like everything they have done, but I understand why they followed a particular strategy to get the collection to where it is. As I've said numerous times, I give them the benefit of the doubt, and am interested to see what they do in future.
  • Vernon:

    I don't think John quoting a contract is «an attack». Maybe he is just making a question about a possible contradiction between the OFL and GWF contract.
    en.wikipedia.org/wiki/Straw_man

    John:
    the wording regarding deliverables and derivatives seems like it could be interpreted as prohibiting this.
    This ambiguity exists if a «Pro» version is a derivative work.
    If there is a certain propietary font (Pro font) and part of it is released under OFL license there is no «prohibition».

    I think, if there is any ambiguity, is hard to claim the enhanced version as derivative when the copyright holder made both.

    In my oppinion is clear that is not the spirit, but I'm not a lawyer.
  • John HudsonJohn Hudson Posts: 2,946
    If there is a certain propietary font (Pro font) and part of it is released under OFL license there is no «prohibition».
    Agreed, that seems unambiguous. The GF contract specifies 'materials created, conceived, prepared, made, discovered or produced by Contractor as part of the Services', so anything that pre-exists the agreement would presumably not be covered by the license requirement. This suggests that font makers who release portions of existing font families under the agreement and font makers who are funded by Google to create works may be in a different legal situation regarding subsequent licensing of 'Pro' versions.

    I focused in my original question on the conjunction of the GF agreement with OFL license, but thinking about it some more, the conjunction of GF and Apache 2.0 isn't that different, in that the GF contract terms re. open licensing of derivatives might be taken to override the permissive nature of the Apache 2.0 license (which allows derivative works to be released under any form of license, including non-open).
    I think, if there is any ambiguity, is hard to claim the enhanced version as derivative when the copyright holder made both.
    I only have a little prior experience with open source licensing, but sometimes the question of whether something is derivative is dependent on the specific code flow. In that case, Ralph Hancock and I released our Biblical Hebrew layout model under the MIT license, and SIL released their fonts using that model under the OFL license (which our MIT license permits them to do). This meant that a subsequent derivative would have different license requirements depending on whether one started with code from me or from SIL.
  • John HudsonJohn Hudson Posts: 2,946
    hard to claim the enhanced version as derivative when the copyright holder made both
    I'm not a lawyer either, but so far as I know the definition of a derivative work in law is entirely based on its relationship to the original work, not on who made it. If I write a novel and then translate it into another language, I would own the copyright on both but the translation would still be classed as a derivative work.

    I wonder if Matthew Butterick is still following this discussion; I suspect he bailed. It's handy to have a lawyer in the room sometimes.
  • The OFL makes clear distinction between "Original Version" (the collection of Font Software components as distributed by the Copyright Holder), and, "Modified Version" (any derivative made by adding to, deleting, or substituting — in part or in whole — any of the components of the Original Version, by changing formats or by porting the Font Software to a
    new environment.)

    A 'derivative' , as commonly framed by libre font licensing, is a 'modified version' of an OFL font made by a license user, under the terms of the license. On the other hand i don't think i've ever heard the term 'derivative' applied to any font made by the license holder. There seems to be no way within the legal framework of e.g. the OFL, for the license holder to make a 'derivative', aka a 'Modified Version' of their 'Original Version'.

    So if i create a 'SlabSerif-Regular', and then you make a bold version of my font, 'SlabSerif-Bold', your bold version is a derivative of my 'Original Version'. If i make 'SlabSerif-Bold' though it is not a 'derivative', it is just another 'Original Version'.

    Looking at the specific wording of the agreement you are stating, i can only really see that it reasonably means the following (in sort of plain english);
    "all the materials used to create the Original Version of the font, and all Modified Versions of the font (e.g. subsetted, autohinted, or auto-generated woff versions, created by google), created under the brief in the contract, shall be made available to the contractor."

    I'm aware that i could be totally wrong :) but i think the wording you have stated is not ambiguous when set in the context of the contracted brief of creating a libre licensed font. For example, if that agreement did not specify that 'derivatives' created as part of the brief should also be available to the contractor, it would cause more ambiguity.

  • Deleted AccountDeleted Account Posts: 739
    edited December 2013
    Isn't needing a lawyer to make libre fonts seems somewhat constricting on the kinds of people who'd be willing to doing it with eyes wide open?
  • vernon adamsvernon adams Posts: 82
    edited December 2013
    I'm assuming that was a tongue-in-cheek rhetorical question ;p
    Anyway this is America, stepping of the sidewalk needs a lawyer.
    but... it's not actually making fonts that requires too much lawyer input. It's publishing them that may need more legal know-how at hand, and making anything under contract, certainly does.
  • Nick ShinnNick Shinn Posts: 2,129
    So much of what anyone does on the internet requires clicking on an Agree button after supposedly reading a massive incomprehensible legal document. Mostly people operate on good faith, sometimes with the help of FAQs.

    Font licence sellers are already explaining licences to purchasers by taking them through an interactive shopping cart process where they are asked first which member of a typeface family they want, which format, and then what circumstances the font will be used in.

    So if the distributor replaces merely reading a long legalese licence by qualifying usage by means of such a “wizard”, then the onus is shifted.

    At the moment, it appears that Google is using the professional, B2B model which assumes both parties have lawyers, rather than the asymmetric Business to Consumer model with interactive interpretation and qualification.
  • Vern asked me:

    > do you mean that Libre fonts in general could do "damage to the 'font industry'"? or specifically that Libre webfonts could do "damage to the font industry"?

    Libre fonts in general. But of course, any libre font may be used as a webfont (though some more easily/effectively than others), and any libre webfont is a libre font in all respects. No?

    > Also, might there be 'Libre' other things that could therefore also do damage to the font industry? such as Libre font engineering tools? or Libre font creation software? or Libre font server software?, or Libre font rendering software?

    > Also, how do you define the 'font industry'. It's a term that we hear and use a lot, but i'm not sure what it actually encompasses. For example, when is something 'part of' the 'font industry' and when is something 'outside' of the font industry? Are Libre webfonts outside the font industry? or part of the font industry?

    I use the phrase "the font industry" as shorthand for the ecosystem of people and companies who create and license fonts and try to make money off that work.

    Those other kinds of software you mention are used by "the font industry" and are closely aligned with it, but distinct. Those other things don't have the same direct impact on demand for fonts or people's willingness to pay for licensing or commissioning fonts. The impact of libre software in those areas on the font industry is likely to be every bit as complicated as the impact of libre fonts, but indirect and possibly diluted.
  • vernon adamsvernon adams Posts: 82
    edited December 2013
    I use the phrase "the font industry" as shorthand for the ecosystem of people and companies who create and license fonts and try to make money off that work.
    Thanks Thomas. Good clarity :)

    So libre fonts are part of the 'font industry', no?
  • Though it is mostly a black and white topic, I think when one binds large things up in simple statements like the one just written, it fails to capture e.g. the sheer breadth of libre, the 390 degree panorama of everyone's opinion, and the vast reaches of tiny strands of the font industry reaching all the way to everywhere one can possibly imagine eyeballs needing it.

    Sure, some libre fonts are part of some people's definition of the font industry, yes.

    As for the question of the legal complexities of libre fonts, I was asking that because of all the twist and shout on derivatives, modifications, and originals, as it applies to the typography formed by the fonts in question.
  • As for the question of the legal complexities of libre fonts, I was asking that because of all the twist and shout on derivatives, modifications, and originals, as it applies to the typography formed by the fonts in question.
    Do you mean; how could the legal status of any contracts to create libre fonts effect the way libre fonts are used in the wild? I think that's a really good question. I thought that was where JHudson was going in the first place, but not sure.
  • As someone who has been buying -- but not making -- type for two decades, one thing that strikes me as someone who is mostly on the other side of the 'font industry' equation is that even though technology has radically changed almost every aspect of type design, production, distribution, use and consumption in the time, I can see no instances of evolution or innovation on the pricing side of the equation. The only major change I've seen is the attempt to move the idea of licensing further down the chain of consumption (charging for every instance in which someone consumes a typeface, via a webpage, ebook, etc.). This isn't dissimilar from the music industry (or film and television), and the net result is likely to be as self-defeating (people will gravitate towards piracy or mediocrity in return for economy).

    I think using Google is a poor proxy for a more complex problem of computing power outflanking a lot of what was considered traditional 'creative' work in many fields, and type being even more constrained by the practical applications/requirements. With a dash of Borges and some techno-utopian cant, one could make the argument that you could legitimately make a clean-room Helvetica (which is a fairly simple casting of the process of derivation and appropriation in art everyone faces at some point, or simply can't because our unconscious eventually occludes all the sources of inspiration). Or simply ratchet down the math to sub pixel logic and claim atomic level differences provide copyright protection.

    Likewise the libre conversation seems a little distracting as well, since the only reason Google has any interest in it is business expediency. Their goals are unambiguous: improving the user experience in the browser generates revenue at a rate that justifies this experiment. If a 'better' (more lucrative in the context of their operations) license comes along, or better technology, GFL could disappear overnight.

    It seems like the underlying issue is that there isn't actually 'a' type industry. In the same way authors thought they were a community when B&N, and Amazon, and Google Books came along, and then learned the hard way that educated individuals that don't have a somewhat arbitrary political arm (union, e.g.) to defend their interests can't come together fast enough to respond to major changes in their economic model, we're seeing the same thing happen in response to GFL: an intervention so massive that it literally reshapes the market, and there's literally no common ground for responding.

    The fact that income distribution among type designers is dramatically uneven means that organization will never happen, since concentration of capital really doesn't unwind due to market forces, and the people who actually profit from designing type will cling to dying models since they can see path of 10-30 years of even declining income (offset by monographs, teaching posts, conference honoraria, etc.) as acceptable. In that sense it's not a broken model, since superstars are always needed and accommodated, but it mostly reflects a severe weakness in American culture: that individuals here mostly reject any sort of collective action in hopes of the lottery pay for the exceptional few.
Sign In or Register to comment.