udi.html (8619B)
1 <!--#include virtual="/server/header.html" --> 2 <!-- Parent-Version: 1.96 --> 3 <!-- This page is derived from /server/standards/boilerplate.html --> 4 <!--#set var="TAGS" value="essays aboutfs free-nonfree" --> 5 <!--#set var="DISABLE_TOP_ADDENDUM" value="yes" --> 6 <title>The Free Software Movement and UDI - GNU Project - Free Software Foundation</title> 7 <!--#include virtual="/philosophy/po/udi.translist" --> 8 <!--#include virtual="/server/banner.html" --> 9 <!--#include virtual="/philosophy/ph-breadcrumb.html" --> 10 <!--GNUN: OUT-OF-DATE NOTICE--> 11 <!--#include virtual="/server/top-addendum.html" --> 12 <div class="article reduced-width"> 13 <h2>The Free Software Movement and UDI</h2> 14 15 <address class="byline">by Richard Stallman</address> 16 17 <p> 18 A project called UDI (Uniform Driver Interface) aims to define a 19 single interface between operating system kernels and device drivers. 20 What should the free software movement make of this idea?</p> 21 <p> 22 If we imagine a number of operating systems and hardware developers, 23 all cooperating on an equal footing, UDI (if technically feasible) 24 would be a very good idea. It would permit us to develop just one 25 driver for any given hardware device, and then all share it. It would 26 enable a higher level of cooperation.</p> 27 <p> 28 When we apply the idea to the actual world, which contains both free 29 software developers seeking cooperation, and proprietary software 30 developers seeking domination, the consequences are very different. 31 No way of using UDI can benefit the free software movement. If it 32 does anything, it will divide and weaken us.</p> 33 <p> 34 If Linux supported UDI, and if we started designing new drivers to 35 communicate with Linux through UDI, what would the consequences be?</p> 36 37 <ul> 38 <li> People could run free GPL-covered Linux drivers with Windows systems. 39 <p> 40 This would help only Windows users; it would do nothing for us users 41 of free operating systems. It would not directly hurt us, either; but 42 the developers of GPL-covered free drivers could be discouraged to see 43 them used in this way, and that would be very bad. It can also be a 44 violation of the GNU GPL to link the drivers into a proprietary 45 kernel. To increase the temptation to do so is asking for trouble.</p></li> 46 47 <li> People could run nonfree Windows drivers 48 on <a href="/gnu/linux-and-gnu.html">GNU/Linux</a> systems. 49 <p> 50 This would not directly affect the range of hardware supported by free 51 software. But indirectly it would tend to decrease the range, by 52 offering a temptation to the millions of GNU/Linux users who have not 53 learned to insist on freedom for its own sake. To the extent that the 54 community began to accept the temptation, we would be moving to using 55 nonfree drivers instead of writing free ones.</p> 56 <p> 57 UDI would not in itself obstruct development of free drivers. So if 58 enough of us rejected the temptation, we could still develop free 59 drivers despite UDI, just as we do without UDI.</p> 60 <p> 61 But why encourage the community to be weaker than it needs to be? Why 62 make unnecessary difficulties for the future of free software? Since 63 UDI does no good for us, it is better to reject UDI.</p></li> 64 </ul> 65 66 <p> 67 Given these consequences, it is no surprise that Intel, a supporter of 68 UDI, has started to “look to the Linux community for help with 69 UDI.” How does a rich and self-seeking company approach a 70 cooperating community? By asking for a handout, of course. They have 71 nothing to lose by asking, and we might be caught off guard and say 72 yes.</p> 73 <p> 74 Cooperation with UDI is not out of the question. We should not label 75 UDI, Intel, or anyone, as a Great Satan. But before we participate in 76 any proposed deal, we must judge it carefully, to make sure it is 77 advantageous for the free software community, not just for proprietary 78 system developers. On this particular issue, that means requiring 79 that cooperation take us a step further along a path that leads to the 80 ultimate goal for free kernels and drivers: supporting <em>all</em> 81 important hardware with free drivers.</p> 82 <p> 83 One way to make a deal a good one could be by modifying the UDI 84 project itself. Eric Raymond has proposed that UDI compliance could 85 require that the driver be free software. That would be ideal, but 86 other alternatives could also work. Just requiring source for the 87 driver to be published, and not a trade secret, could do the 88 job—because even if that driver is not free, it would at least 89 tell us what we need to know to write a free driver.</p> 90 <p> 91 Intel could also do something outside of UDI to help the free software 92 community solve this problem. For example, there may be some sort of 93 certification that hardware developers seek, that Intel plays a role 94 in granting. If so, Intel could agree to make certification more 95 difficult if the hardware specs are secret. That might not be a 96 complete solution to the problem, but it could help quite a bit.</p> 97 <p> 98 One difficulty with any deal with Intel about UDI is that we would do 99 our part for Intel at the beginning, but Intel's payback would extend 100 over a long time. In effect, we would be extending credit to Intel. 101 But would Intel continue to repay its loan? Probably yes, if we get 102 it in writing and there are no loopholes; otherwise, we can't count on 103 it. Corporations are notoriously untrustworthy; the people we are 104 dealing with may have integrity, but they could be overruled from 105 above, or even replaced at any time with different people. Even a CEO 106 who owns most of the stock can be replaced through a buy-out. When 107 making a deal with a corporation, always get a binding commitment in 108 writing.</p> 109 <p> 110 It does not seem likely that Intel would offer a deal that gives us 111 what we need. In fact, UDI seems designed to make it easier to keep 112 specifications secret.</p> 113 <p> 114 Still, there is no harm in keeping the door unlocked, as long as we 115 are careful about who we let in.</p> 116 </div> 117 118 </div><!-- for id="content", starts in the include above --> 119 <!--#include virtual="/server/footer.html" --> 120 <div id="footer" role="contentinfo"> 121 <div class="unprintable"> 122 123 <p>Please send general FSF & GNU inquiries to <a 124 href="mailto:gnu@gnu.org"><gnu@gnu.org></a>. There are also <a 125 href="/contact/">other ways to contact</a> the FSF. Broken links and other 126 corrections or suggestions can be sent to <a 127 href="mailto:webmasters@gnu.org"><webmasters@gnu.org></a>.</p> 128 129 <p><!-- TRANSLATORS: Ignore the original text in this paragraph, 130 replace it with the translation of these two: 131 132 We work hard and do our best to provide accurate, good quality 133 translations. However, we are not exempt from imperfection. 134 Please send your comments and general suggestions in this regard 135 to <a href="mailto:web-translators@gnu.org"> 136 <web-translators@gnu.org></a>.</p> 137 138 <p>For information on coordinating and contributing translations of 139 our web pages, see <a 140 href="/server/standards/README.translations.html">Translations 141 README</a>. --> 142 Please see the <a 143 href="/server/standards/README.translations.html">Translations README</a> for 144 information on coordinating and contributing translations of this article.</p> 145 </div> 146 147 <!-- Regarding copyright, in general, standalone pages (as opposed to 148 files generated as part of manuals) on the GNU web server should 149 be under CC BY-ND 4.0. Please do NOT change or remove this 150 without talking with the webmasters or licensing team first. 151 Please make sure the copyright date is consistent with the 152 document. For web pages, it is ok to list just the latest year the 153 document was modified, or published. 154 155 If you wish to list earlier years, that is ok too. 156 Either "2001, 2002, 2003" or "2001-2003" are ok for specifying 157 years, as long as each year in the range is in fact a copyrightable 158 year, i.e., a year in which the document was published (including 159 being publicly visible on the web or in a revision control system). 160 161 There is more detail about copyright years in the GNU Maintainers 162 Information document, www.gnu.org/prep/maintain. --> 163 164 <p>Copyright © 1998, 2021 Richard Stallman</p> 165 166 <p>This page is licensed under a <a rel="license" 167 href="http://creativecommons.org/licenses/by-nd/4.0/">Creative 168 Commons Attribution-NoDerivatives 4.0 International License</a>.</p> 169 170 <!--#include virtual="/server/bottom-notes.html" --> 171 172 <p class="unprintable">Updated: 173 <!-- timestamp start --> 174 $Date: 2021/10/01 17:02:54 $ 175 <!-- timestamp end --> 176 </p> 177 </div> 178 </div><!-- for class="inner", starts in the banner include --> 179 </body> 180 </html>