Phone, Tablet and Gizmo Thread #0x02

Raja Bose · Post by **Raja Bose** » 02 May 2013 07:22

Mort Walker wrote:
Raja Bose wrote:Saying HW is commoditized doesn't mean its easy to do - thinking so is a fallacy. It just means one's profit margins on selling HW onlee are much lower and that is pretty much true for everybody except FruitCo. Just do a calculation on how many units of phones Sammy needs to sell to make a million $ in profits (with capital costs amortized) vs someone selling units of SW like Mickey and you will get the answer. Value addition which results in customers ponying up more cash rarely happens at the HW level now hence, RoI is lower. Is mango abdul customer going to pay me $20.- more if my GPU does extra FLOPS or will he pay me more if my camera software can automatically push pics to Instagram? Add to this the high capital cost of doing HW R&D, development, tooling and not to mention the risk and its evident why being HW-onlee is not sustainable for anybody in the CE business. This is another reason why VCs in the valley shy away from investing in HW startups - too high risk, too low RoI.
This is what I was trying to get at earlier. We are already at a point where H/W development is done with S/W tools as we have FPGA, SoCs, NoCs, and S/W defined radio. However, the big differentiation in devices will not come from S/W alone, but by fundamental breakthroughs in the basic sciences. If we see materials with high dielectric constant, super capacitors, batteries with nano-tube structure that see a huge increase the surface area per volume - we'll have the next breakthrough for CE devices. This stuff is going on in the defense and aerospace industries, but it's economies of scale. There simply aren't enough of these devices to make it to the CE world yet. This is where unkil makes great strides by participating in the space race, the cold war or any wars!

Saar what you are referring to is a long-term thing and is differentiation at a much more fundamental level. There is no doubt that will happen and like before a lot of it will come from defence driven research (example, wearable display technology). Product centric companies are more short-term focused and for them the key is how to differentiate now with lowest amount of risk, rather than 10-15-20 years later. The companies with deep pockets and strong R&D programmes will continue to do the basic research and push the envelope but that is only a handful of companies. And any such breakthrough in HW provides a very temporary time advantage to the original innovator (if any at all) before the rest of the industry grabs it and makes it a commodity. Becoz then hordes of chipanda, taiwanese and koreans will jump in to make it at 10% of the cost leaving not even enough RoI for the original inventor to recoup research costs. In fact in most cases the original innovator is left sucking his thumb becoz someone else sees a better value-added way to package it into a product for the consumer. Like the Mahdi said, "perception is everything". If you ask the random abdul on the street, they will go "ZOMG dont you know?!!!! Apple invented multi-touch and the Mahdi personally dreamed it up while smoking crack thru his mush crack!!!! INNOVASHUNNNNNN!!!!!". None of them have heard of Synaptics, Sharp, PARC, PerceptivePixel ityadi. Recent example: The sooper-dooper capactivie touchscreen controller which Synaptics and GB after years of research, unveiled last year in the Lumia 920 - works with gloves, works with fingers, works with a stick (and a d1ck)...guess what? Now local chipanda manufacturers are selling touch controllers with the same feature for 60% of the price. That is commoditization.

Vayutuvan · Post by **Vayutuvan** » 02 May 2013 08:23

Raja Bose wrote:^^^No worries there - I know Anujan-ullah IRL. Sorry to burst your betting pool.

Well, I don't gamble often. That said, had you "fixed" the match you could have collected a few K "khota sikka" for your blur pad. Missed your chance.

of course goes without saying that TFTAs posing as SDREs would hardly miss a few measly khota sikka.

So, IRL is [International|Internal] Resource Locator, [Internal|Intermediate] Representation Language or ...? I have limited imagination and feeling a bit lazy to ask chacha.

Raja Bose · Post by **Raja Bose** » 02 May 2013 09:32

^^^Nothing as complicated. IRL == In Real Life.

Vayutuvan · Post by **Vayutuvan** » 02 May 2013 09:40

In case it was ambiguous I meant it is you who is the TFTA in SDRE clothing. Just sayin'.

Raja Bose · Post by **Raja Bose** » 02 May 2013 10:23

I am boor SDRE onlee - Even my last 40 generations were not even close to TFTA and neither will the next 40 generations (if I decide to procreate i.e.).

member_20292 · Post by **member_20292** » 02 May 2013 12:53

Raja Bose wrote: but by fundamental breakthroughs in the basic sciences. If we see materials with high dielectric constant, super capacitors, batteries with nano-tube structure that see a huge increase the surface area per volume - we'll have the next breakthrough for CE devices.

As an expert in this field ....please don't hold your breath.

Vayutuvan · Post by **Vayutuvan** » 02 May 2013 18:23

Raja Bose wrote:I am boor SDRE onlee - Even my last 40 generations were not even close to TFTA and neither will the next 40 generations (if I decide to procreate i.e.).

Bose babu, in our town the word is that lots of Californians are coming in with bags of money and buying up all the real-estate. The bags of money is from FB IPO and royalties from mobile patents.

Raja Bose · Post by **Raja Bose** » 02 May 2013 21:48

^^FB IPO was a bust for most FB employees especially those who joined after 2008 or so. They went from TFTA to phata abdul pronto. FB is growing slow and steady so they might get back to TFTA eventually maybe. BTW did FB make more profit than Takla this quarter?

Raja Bose · Post by **Raja Bose** » 02 May 2013 23:51

Company gets the royalty then they calculate what % an employee gets based on a formula. Compared to what the company earns, the employee gets peanuts. But its not a bad sum if looked at in isolation from the context. Ofcourse not all companies share this royalty and esp Khan companies will just point to the 1-time sum they paid you when they filed the patent as the reward you get for the invention. That 1-time sum also is less generous in Khan companies compared to say Oiroes and a lot of Khan companies will further slice and dice that 1-time sum into smaller payments - 1 during filing, 1 during grant etc. As a result money cannot be a motivator for most inventors employed in a company. On the flip side if you have filed patents as part of your startup and they catch the fancy of some big company who has critical need for it, you can easily get $250K-$1 million for each patent.

Vayutuvan · Post by **Vayutuvan** » 03 May 2013 01:22

I am talking of patents like CDMA (and Viterbi coding - has this patent run out?) of Qualcomm and the original RSA patent . If somebody is working as an employee of a big corp., then it is more like a paper publication. The difference is that paper details are available for everybody and each can have their own implementation where as patent is for one such implementation. IP rights protection can take the form of copyright (for software), patent (for process/chip/gadget/device) or company secret. With the first two, the information is public and can be examined by anybody (once they pay the fee to USPTO). With copyright, if the copyright holder wants to be notified every time somebody looks at the copyright, then they have to pay some fee (IIIRC $100 or some such) per incident. For software copyright, one can deposit either one or both of binary/source code. In that respect, copyright is totally useless. If somebody reverse engineers the binary and rewrites for another architecture in some other language and compiles with some compiler with different switches, there is no way anybody can show that it was copied. If it the source, they can examine the critical parts of the code would be able to re-implement in a clean room environment. Usually for most algorithms, the general principles are well known mathematical formulas/results. What makes the critical difference is as follows:

1. Most high-performance algorithms are parameterized using a small set of parameters. Because of that the algorithm is not one single algorithm but is a family of algorithms. For example, if the algorithm is parameterized using 4 parameter each having say 4 settings on the average, one would end up with 4^4 or 256 different algorithms which belong to the same family. Sometimes these settings are not discrete values but ranges of values whihch depend on the problem.

2. The parameters are selected (or even new ones introduced) depending on the amount information known about the problem that is being solved. Some are selected once for all for a particular problem domain, others have to be tuned depending on the problems one would see for that particular application. Some others are dependent on machine architecture and track the technology - say the number of FP pipelines, pipeline lengths, interconnect, number of processors/cores, cache size. NUMA etc.

3. Choosing the parameters requires quite a bit of knowledge about the over algorithm and all the sub-problems that are being solved along with the problem domain peculiarities.

So essentially a competitor who wants the tech can take a look at the source code and find out how the overall problem solving strategy has been parameterized and the typical ranges for their particular domain. That's all they need to know because one has to assume that they are as competent as you are and know the problem domain as well as, may be even better than, you do. Some of them might have even better market information (i.e. the near future trajectory of the problem domain). After that it is a matter of time for them to get their own implementation. Only thing that prevents established companies going through this process of re-implementation is that there is always the route of licensing the technology or acquire the company if it is a start-up. Other business pressures like whether they are losing market share and if so at what rate, how much it costs to reimplement etc. also dictate (in most case it is nothing but a make or buy decision that needs to be made) their course of action. Of course, all bets are off If the other is as big or a bigger company and is not willing to license/sell.

So, for start-up companies with new and unique technology, the decision to copyright, patent, or keep it a company secret is an important decision and must be made with some caution.

Raja Bose · Post by **Raja Bose** » 03 May 2013 06:33

Why would you need to pay USPTO to look at the contents a patent hain ji?

Just go to uspto.gov and do a full database search. Once a patent is published (18 months after filing of utility application), the contents are public for anyone to see. Before that no one except inventor or authorized agent can see anyways. The path I prefer to take is to file provisional immediately to get the earliest priority date possible even if application is not fully threshed out. Then work on the utility patent application in that 1 year and do magazmaari on the claims ityadi. For startups this strat-e-jee helps from a cost perspective too since USPTO only charges small entities a relatively small fee to file a provisional and any abdul can do it online within minutes.

---

In a startup, copyright is as useful as Manmohan's Sharm-el-Sheik agreement. If a startup keeps a trade secret instead of filing a patent, it should do so only if it can build a product with tangible benefits out of it, can demonstrate the benefits to consumers/investors and is confident that without being revealed nobody can look at the product and reverse engineer the technology. One cannot defend trade secrets from infringement unlike patents (unless it resulted from violation of some signed agreement) and one cannot leverage them as assets during a sale unless above stuff is ensured.

Vayutuvan · Post by **Vayutuvan** » 03 May 2013 07:00

RB I am talking about copyrights where the code is deposited with ... (is it USPTO?). In any case, if the CR holder submits only hardcopies, then who ever wants to look at it has to make a trip to district of chor-e[r][r]-columbia to do that which I am sure would cost much more than $100 I quoted above. The point is that anybody and her aunty can take a look as long as they are willing to take the trouble. Company secret is different from trade secret. In the former, one doesn't have to deposit the code/process/<x> with any body - all they have to do is to write it up and place it in a locker (or not even do it, but obviously have to bear the risk of losing the tech altogether in case of a catastrophe like plane crash). The risk of course is that if somebody reproduces the tech almost nothing can be done ("almost" because one should hold back a la recessed deterrent, you see), but then there are risks with every course of action. The usual way company secrets get out are on "two legs", i.e. people working on the product are hired away by competitors. I know the objection to that is "what about non-compete?". Well, usually it is hard to prove that the tech had been copied - especially if the competitor makes a big show of "clean room tech". People who know have the knowledge and would use it in the new job and say that I have not used it. Who can disprove?

Raja Bose · Post by **Raja Bose** » 03 May 2013 07:31

What you defined above is a trade secret. Dunno what is company secret. Copyright is pretty useless as a protection unless its for literary works/works of arts ityadi I guess.

The loophole used for people walking out with IP in their heads is that they should not "consciously use" knowledge from old employer. Ofcourse there is almost no effing way of proving that they have consciously applied knowledge from their old employer. So its all pretty much a farce.

Even in Chacha's case where they did find some src files with Sun Microsystems copyright notices in the code, the punishment was pretty mild. And re. no-compete clauses in hiring agreements, in people's republic of California no-compete is not enforceable so companies cant do jack if someone just waltzes out.

Vayutuvan · Post by **Vayutuvan** » 03 May 2013 09:18

What I remember from a book (published by IE^3) more than two decades back is that a trade secret can be recorded and escrowed which is not publicly accessible but can be checked when disputes arise where as a company secret is not registered with any third party so it offers no legal recourse at all in case of violations.

To conclude there are four broad ways of protecting IP and each has it's pros and cons. Obviously that's why you see all these lawyers from sonsini blah showing off their tech chops and charge at least 500 per hour for their undeniable upper hand in IP law. Ultimately most of the IP is junk so is most of the patents /copyrights/trade secrets. As per trademarks China thumbs their nose at rest of the world with things like "Polystation" consoles and what not. It is all a hit or a miss, being at the right place at the right time and proffer one's expertise degree, connections etc. A lot of ideas are going to die in the lab after some initial excitement and a few million going from VCs pockets to the employee pockets. if the startup goes under, employees and VCs regroup under different umbrellas and try yet another next "big" thing. Makes the world go round.

Raja Bose · Post by **Raja Bose** » 03 May 2013 09:37

Trade secret doesn't need to be stored with some escrow entity and is essentially the same as what you are describing as company secret (i.e. company secret in escrow != trade secret). Trade secrets are not available for public inspection unless it is part of discovery of some other violation (such as patent infringment). In that case you cannot plead the 5th on your secrets.

But it is incumbent upon the plaintiff to show why should some company's trade secret be available for inspection (design docs, e-mails, src etc.) and in most cases such requests might get thrown out unless judge or jury is biased (like FruitCo vs Sammy case). Patent law is a great field for ITvity injineer types who want to make more khota sikkas than possible with regular ITvity.

Satya_anveshi · Post by **Satya_anveshi** » 04 May 2013 03:16

interesting perspective coming after the market share report that bose boss previously posted (Nexus7 not selling very many units):
Big phones, phablets & tablets account for just 10% of Android's installed base

Vayutuvan · Post by **Vayutuvan** » 04 May 2013 04:44

RB

>> Trade secrets are not available for public inspection unless it is part of discovery of some other violation (such as
>> patent infringment). In that case you cannot plead the 5th on your secrets. But it is incumbent upon the plaintiff to
>> show why should some company's trade secret be available for inspection (design docs, e-mails, src etc.)

My point is the other way round that if employee(s) from TS holder goes(go) to a competitor or anybody working/consulting for the competitor came in contact with TS holders' tech and subsequently the competitor develops a competing product or files for patent for that tech, then the TS holder can prove that the tech came from their trade secret that has been made available on a need-to-know basis to those people under some kind of an NDA and can sue the individuals and the competitor in case they were told about the NDA but went ahead nevertheless. For this to happen, the TS holder should have the TS escrowed which they can make it known only to arbitrators/experts.

Raja Bose · Post by **Raja Bose** » 04 May 2013 05:15

^^^Actually that's not the case. They can provide the engineering/research notebooks/design docs/any other forms of communication to prove their case and will have to make it available to the other party's lawyers too - escrow doesn't save them there. It is not incumbent upon the IP owner to put their trade secret information in escrow in order to prove infringement. In fact it is standard SOP in pretty much all ITvity companies at least that one should never discuss patentability of an idea, comparison with competitors etc. over e-mail or any other form of written communication. Becoz all that can be used in court. But any inventor is encouraged to keep research notebooks which are turned over to the company when they leave where a record is kept of when the inventor got the idea, how it evolved into the idea for which protection was filed for etc.

In reality the plaintiff will try to only provide the minimal evidence required to prove their allegation and in a lot of cases they won't even pursue the case becoz then the case becomes like a bikini on the plaintiff - it reveals more than it hides (as our Sidhu paaji would say). That is why patents are used and companies with strong patent portfolios are able to use it to their advantage (IBM, ChipZ, Mickey, GB). Just a day or so ago I was having a chat with a Chacha BD person and both Chacha and Sammy are very very paranoid about getting sued for IP infringement now that they have put in place all sorts of checks to prevent taint. Sammy was like that since 2011 or so but I guess in case of Chacha, it is more recent especially after Motor Oil's IP portfolio proved to be pretty much a dud (as yours truly had mentioned on this dhaaga back when Chacha pay $12B to buy Motor Oil

).

SaiK · Post by **SaiK** » 05 May 2013 02:15

Can those lumia 4g devices, take 2g cards? say I get the device unlocked from AT&T? The question is basically towards using the device while travelling to desh. Wife is travelling, and I have a 2G sim card. Can I buy into a plan with ATT or GSM providers, and get it unlocked+use the card/temp use? Need a new plan btw..

Vayutuvan · Post by **Vayutuvan** » 05 May 2013 04:18

RB moshay, I will not revisit RB-vina samvad. So I will summarize now and let us hold our peace for ever on this issue (and IANAL and dredging from 20 year old memory). We agree on all points except

You contend that Trade Secret == Company Secret

as against

my contention that there is a subtle difference and the difference is only limited to (because that is all I know and remember) protection of software/programs excluding embedded, hardware, ASICs)

Let us move on until one of us has more info one way or other.

Raja Bose · Post by **Raja Bose** » 05 May 2013 05:33

^^^Saar at least to my knowledge I have never even come across the term company secret being used whether internally or during prosecution (in USPTO or in court). I may be wrong ofcourse or it may actually have been in use in IP law many moons ago but is no longer so.

Now 2 examples. A good example of trade secret is the Coca Cola original formula. That is a trade secret even though it is vulnerable to reverse engineering and has been reverse engineered many times. If such a benefit as what you defined in trade secret was available Coca Cola would have availed of it. And another data point I can provide you is the case of my BIL where many of his inventions in Motor Oil were deemed trade secret instead of being patented. Those were typically specific IC designs or optimizations in design (for baseband processors, DSPs for voice and other mundane stuff). Motor Oil would have been willing to apply for protection such as what you defined as trade secret since chipandas reverse engineered pretty much all those designs but they can't. In fact for 2-3 of those trade secrets some see-near Motor Oil patent fella got mush kicked for making the wrong call since it was determined that critical parts of those inventions should have been deemed patenteable in which case some protection would be available for Motor Oil to fight back (at least in US/Europe since chipanda laws for patent protection as as lax as paki abduls using protekshun). Making it a trade secret left no recourse for Motor Oil except security thru obscurity.

Vayutuvan · Post by **Vayutuvan** » 05 May 2013 06:17

It is essentially "security through obscurity".

Oh. and cease fire

Raja Bose · Post by **Raja Bose** » 05 May 2013 10:08

chaanakya · Post by **chaanakya** » 05 May 2013 10:18

That looks like PeecHaDdi or PakicHaDdi

chaanakya · Post by **chaanakya** » 05 May 2013 10:27

PakicHaDdi holders are here.

Vayutuvan · Post by **Vayutuvan** » 05 May 2013 22:24

It is peeelachaddi (AKA pakichaddi), not a towel. If I were to throw mine it would be pristine white.

I got off of my virtual backside and finally googled.

Purely going by Google results and a cursory reading of an article in Harvard journal of law and tech., there seems to be no difference between "Trade Secret" and "Company Secret". In fact there is no term like "Company Secret".

Now that's out of the way, the wider point I was about to make is this. There is no effective protection for software. One can give binaries only but it is not always convenient to the licensees, especially if the licensee has QoS/maintenance contracts with their customers. In that case they would have to fix within a certain time period. If the license granter is a start-up or a relatively small company, then the licensee would press for source code so that they can be fix any problems if they have to meet some deadlines. It behoove the license granter to take this into account during license negotiation process. Binaries are also not convenient because licensee usually has a large customer base who uses various OS version/architecture platforms. In the interest of keeping the operating costs low a start-ups usually would not maintain all the different platforms and configurations. For a software start-up, it is very important to stick to the standards - de jure or de facto - as far as possible.

One way out of this conundrum is to open source the code, not necessarily under GPL or LGPL, but under ones own custom license. The custom license should include language to the effect that what is being transferred is a "trade secret" which allows the granter to keep the IP rights - publication rights, claims of credit, and further extensions and derivative works. Furthermore the language should include clauses that require the licensees to protect the IP as they protect their own "trade secrets" or better - for example need to know, NDAs. This is all with an eye towards delaying the inevitable, i.e. the technology getting out.

Raja Bose · Post by **Raja Bose** » 06 May 2013 00:15

^^^So you meekly submit to my upper hand in all takniki matters like Salala ghazis did to NATO?

Software has no protekshun and also has the most ridiculous protections (such as FruitCo's pinch-zoom gesture patent).

Raja Bose · Post by **Raja Bose** » 06 May 2013 20:56

Motor Oil lands in further Hot Oil

Poor Motor Oil doesn't seem to catch a break and seems to be throwing punches like keshto.

Anyhow they will prove their worth once they are used for their true purpose i.e. Chacha going vertical on devices as opposed to their stated purpose ("strong moth-ridden patent portfolio to defend Android").

Vayutuvan · Post by **Vayutuvan** » 06 May 2013 21:53

Raja Bose wrote:^^^So you meekly submit to my upper hand in all takniki matters like Salala ghazis did to NATO?

On a contingency basis, i.e. till I, along with my harried tired and beaten forces, retreat to my deep dark kingdom across the seven seas. But watch out for UAV strikes out of the blue yonder.

Raja Bose · Post by **Raja Bose** » 06 May 2013 23:07

^^You might want to try Abbottabad guest house for R&R, I hear its more comfy than slumming it in Tora Bora.

ManuJ · Post by **ManuJ** » 07 May 2013 06:07

Raja Bose and matrimc, thanks for the discussion on protecting IP.
As someone who's in the planning process of getting a start-up going, this is good info.

I am still not very clear about what makes a particular piece of software patentable.
If I create a new online game, is the game playing strategy patentable?
If I create an AI agent, is that patentable?
If I come up with unique s/w architecture for my product, is that patentable?

Raja Bose · Post by **Raja Bose** » 07 May 2013 06:42

Patentability depends on first the novelty of the invention (i.e. not obvious to "a person skilled in the arts") and second, the articulation of inventive steps. The latter also includes a novel combination of existing known steps to do something which is non-obvious to a person who is an expert in that field (aka "a person skilled in the arts"). You can file for patent for just about anything but you have to think about:
1) Whether it is patentable based on above criteria?

2) Even if (1) is true, can it be defended i.e. can you detect if someone is violating your patent and how easy is it to detect such a violation *reliably*?

3) Even if (1) and (2) are true, is it worth defending? Here you can look at two things: (a) Protection: Will your patent prevent somebody else from either gaining competitive advantage over you or prevent them from handicapping your business/source of revenue/product's USP etc. (b) Is there a chance that now or in the future some one will violate your patent? Becoz if your patent is too restrictive then people might simply find ways to circumvent it without violating it. In that case its kinda useless.

4) Patent is kinda like granular locking taught in CS. What is the granularity at which you want to lock down? Keep it too generic and it wont be prosecutable even if granted by USPTO/under PCT. Keep it too specific and people will just sidestep what you have protected and avoid violating your precious patent.

Patent filing is expensive, time consuming and a long drawn process (average filing to grant is ~3-5 years if aal is well) so for a startup these things require much higher consideration when deciding whether to file. Filing for vanity's sake is not advisable. Also when you file as part of a startup, you need to be careful about assignment rights which transfers ownership of that invention to the company as opposed to individual inventors. Sometimes if co-founders are not getting along well, individual inventors can withhold assignment and use it as a bargaining chip (unless there is a pre-existing employment agreement they have signed which requires automatic assignment in which case you will have already signed and executed the assignment document like you do as part of employment at a big company). And during acquisition, most likely an acquirer will not buy any patents which is not assigned to the company being acquired.

Vayutuvan · Post by **Vayutuvan** » 07 May 2013 07:01

Also consider the following:

1. Patent runs out after 20 years.
2. During the patent process, search for "prior art" is carried out. It is better to DIY and if you find prior art, then it is not patentable. If you ask an attorney to do it, you are down $500/hr which you can save if it is a little bit of digging through journals/patent database/Google/Google scholar. You and your trusted friends/well-wishers need to act as the first filter. Also, self-criticism is very important in that do not get carried away.

Raja Bose · Post by **Raja Bose** » 07 May 2013 07:15

Or rather if you find prior art (and 99% of the time you will), figure out how to circumvent them.

Vayutuvan · Post by **Vayutuvan** » 07 May 2013 07:26

True that. But the defensibility quotient goes down, right?

Raja Bose · Post by **Raja Bose** » 07 May 2013 07:59

^^^Nope it doesn't. If anything the more well researched your prior art is and claims framed accordingly, the harder it is for the examiner to throw your application back at you.

manish · Post by **manish** » 07 May 2013 08:47

Signs of a (minor) turnaround in HTC's fortunes?

Looks like the early positive response to the HTC One is helping their cause and stemming the bleeding a bit.

Boosted by the One launch HTC revenues rise 23% in April

HTC hasn't had the best of times in the market over the past year and a half, losing huge chunks of its market share and seeing revenues contract rapidly. However, the April numbers show that the company may finally be starting to see the light at the end of the tunnel.

Boosted by the launch of the One flagship, HTC revenues rose 23% in March and reached NT$19.6 billion ($657 million). However, such was the scale of the HTC recent troubles that even after such significant jump the company's earnings are 36.9% lower than they were in the same month of last year. Still the decline was 48.6% last month, so things are at least moving in the right direction.

Raja Bose · Post by **Raja Bose** » 07 May 2013 09:56

Too bad anmol mian can't see this

, though Mortullah take note since it goes against your prediction.

Win8 sales trend is same as Win7 trend: Windows 8 passes 100 million license sales, Microsoft reflects on the six month stage

ManuJ · Post by **ManuJ** » 07 May 2013 10:28

Raja Bose and matrimc, thanks once again.
Got a lot of doubts cleared.

ramana · Post by **ramana** » 07 May 2013 22:59

Matrimc, Are you confusing company secret with proprietary information which is protected from unauthorised disclosure? PI is protected by NDAs and what not. And one can't disclose Thir Party PI without written authorization to disclose even among different projects or programs.