Text why pro­pri­etary or no hard­ware docs hurt the manufacturer

I stum­bled about a text which describes why it is ben­e­fi­cial to dis­close hard­ware pro­gram­ming docs and why it does­n’t help in keep­ing this infor­ma­tion away from the com­pe­ti­tion. I don’t repeat it here, so go and read it.

It’s a lit­tle bit old (last mod­i­fied in 2003), but IMO still up-to-date. If some­one approach­es a com­pa­ny for hard­ware docs, please pro­vide this link to them!

Unfor­tu­nate­ly it fails to men­tion that it would even be nice to get docs for obso­lete or not sup­port­ed any­more hard­ware (if your com­pe­ti­tion learns even stuff from your hard­ware which is 3 – 4 gen­er­a­tions old, it is not real­ly a com­pe­ti­tion and you most prob­a­bly are lead­ing because of inno­va­tion, if not you either are too expen­sive and open­ing the docs would be a rea­son to buy regard­less, or your soft­ware devel­op­ment is not good enough and open­ing the docs would allow users to fix this prob­lem them­selves). This could be a first step for a com­pa­ny to “test the water”. It would be an invest­ment with­out any mon­ey in return (the com­pa­ny does­n’t sell such hard­ware any­more), but it would show the com­pa­ny how it affects their image, how much they have to invest and what they can get in return (when peo­ple do cre­ative things with your obso­lete hard­ware you haven’t imag­ined before, you can bet they can do the same with your cur­rent hard­ware too… you may get an entire­ly new mar­ket “for free”).

If you apply some more thoughts about this top­ic and for exam­ple graph­ic cards, you even notice that any infor­ma­tion the com­pe­ti­tion may get by look­ing at freely avail­able hard­ware docs for graph­ic cards (instead of reverse engi­neer­ing it), can only be used 2 – 3 inno­va­tion cycles lat­er. This is caused by the short turn around times between new graph­ic cards. When a new graph­ic card hits the mar­ket, a devel­op­ment team already works at the sec­ond next gen­er­a­tion (and the next gen­er­a­tion is most prob­a­bly not only in fea­ture freeze but at the bug fix­ing and per­for­mance enhance­ment step). Now, how much val­ue does the com­pe­ti­tion gain from this? I would say only the mon­ey need­ed for the reverse engi­neer­ing. At the same time you gain mon­ey from hard­ware sales from those peo­ple which use (the result of) your hard­ware docs. And the com­pe­ti­tion is required to open their docs too (see below for the “com­put­er freaks” part), so you can safe the mon­ey for the reverse engi­neer­ing lat­er too.

For sound­cards this is a lit­tle bit dif­fer­ent. There you don’t have such short cycles, but cur­rent­ly there you have a pub­lished stan­dard (HDA) and you have Cre­ative with no docs at all on the oth­er side. Hey, Cre­ative, if you stum­ble upon this, what about kick­ing Microsoft in the ass by pro­vid­ing your hard­ware doc­u­men­ta­tion to any­one and ben­e­fit­ing from a lot of peo­ple which are pissed off because their shiny Creative-gear does­n’t work on Vista? I’m sure a lot of peo­ple are will­ing to spend their free time to find a way to make your hard­ware use­able on Vista (and on oth­er OS’) with­out get­ting mon­ey from you. And I’m sure peo­ple will find a way to get stuff out of your hard­ware which makes your eyes fall out of your head (and increas­es hard­ware sales). Oh… yes… hey, VIA, what about the docs for your soundgear too? There’s no mar­ket for sell­ing hard­ware docs, but a huge mar­ket to sell sound hard­ware. And those peo­ple which play around with non-mainstream soft­ware are those peo­ple (com­put­er freaks) which rec­om­mend hard­ware to peo­ple (mom, dad, neigh­bors, friends) which don’t play around but just use main­stream soft­ware. Those “ordi­nary” peo­ple may not depend on your hard­ware docs, but the com­put­er freaks will more like­ly rec­om­mend stuff which works not only on the main­stream stuff (just in case some­one wants to try some non-mainstream stuff).

The same (com­put­er freaks rec­om­mend­ing hard­ware) is true for cable TV / satel­lite TV / … stuff.