Maps Navigation : Google vs a Startup | Hacker News Reader

Maps Navigation : Google vs a Startup

silicontryst.wordpress.com

46 pointsnishantmodak37 comments

This reminded me of google advertising about 'Chrome by google' on big billboards in London. My first thought was 'they are not giving credit to everyone else who did the work on chrome'. Maybe it should have been 'Chrome - an open source project by google'.

There is no advertising clause, but claiming that it is by google is still a big lie considering big parts were not written or designed by them. Including the main part - webkit/khtml, the html renderer.

Especially with open source projects, getting credit for your work may be the only thing you get. Giving credit is a gentle(wo)mans thing to do.

In academia this would be plagiarism... but in business, it seems to be business as usual. Business can get in trouble for giving credit, but academia gets in trouble for not giving credit.

Do you feel the same way about "Mac OS X by Apple", considering how much of that operating system (FreeBSD, Darwin, GNU, KHTML, llvm etc) started life in the open source community?

I'm as big an open source booster as anyone, but I can't say it upsets me in the slightest to see Google taking credit for Chrome in adverts. If anything, it makes me feel like the open source model has won - it's so widespread now it isn't even noteworthy.

Yeah, I do feel the same way about Apple. Apple claims credit a lot more than Google. Google is definitely way more free software friendly than Apple in my experience - like a whole different magnitude of more friendly :)

I see your point, but without giving proper credit, the developers of open source projects will not get credit outside of the project. They won't get the credit from the press or from The People.

Even in other open source projects it would be nice to give credit more evenly. As an example, many interviews or articles mention 'Guido the creator of python'. Whereas the reality is that thousands of people have contributed to python to make it what it is today.

It's hard to give credit even when you try your hardest to. Also, where do we draw the line? Do we start with crediting Persian, Egyptian, and Greek mathematicians and artists? Do we need to give a whole raft of references each time a trumpet is blown? Do we list individual contributors? Or just major contributors, or founders? Or perhaps not mentioning anyone at all - which many hackers do. An 'open source community project' is a nice compromise as it gives credit to everyone.

Perhaps 'Made by the people of the world, for the people of the world' would have been a better advertisement.

-- merry xmas

I think I'd argue that this _is_ analogous to academia (well at least to mathematics). Take Fermat's Last Theorem for example, it's commonly accepted that Andrew Wiles proved the Theorem. However, without the work and results of thousands of other mathematicians throughout history he almost certainly could not have done so. The credit to those people is given within the proof; "combining the results of X with Y's theorem..." not in an enormous list on the front page. Equally, the name Google goes on the advertisements but within source code and blogs and bug fixes part of the credit goes to the technology that Chrome is built on.

yeah, nice point. The standard way to give credit in academia is to list references throughout a work.

It's funny that on the proof it reads - 'By ANDREW WILES*'. So even there he was giving credit. It also mentions Fermat in the title, and within the first paragraph references a bunch of others.

Many open source projects use a section in their README, or a CREDITS file, or a credits link on their web pages... or a 'powered by X'. Many do all of these things.

Web pages often have a tiny link with 'made by X, designed by X, or powered by X'.

The business world often does not have this credit giving culture. Some times you might see 'coffee beans roasted by X' or other little notes on shop windows. Perhaps the employee of the month is another example.

I guess the main point of the original poster is that they would have liked to be given some credit for their work... and that they felt sad and a little depressed to see google taking their credit. They want people to know that they made that and not google - despite what the google marketing machine is telling people.

-- merry xmas

The people that care know that and those who don't know don't care.

I think that if an entity does more than about 99% of the work it takes to bring something into being, it's fine for that entity to take entire credit for it.

about:credits

Are you saying that the authors of libraries used should be counted in the credits, or are you saying that Google is going above and beyond the call of duty? If the latter, I applaud you.

But if you believe Google is obliged to credit those groups in any way beyond the letter of their licenses (e.g. share credit during advertising, as we're discussing), you've got it wrong. Do you also believe that Intel and AMD, Dell and Apple, all of the implementers of browsers that came before Chrome, and everyone down to Djikstra should be credited? Because, you know, they contributed in some way to Chrome's existence too. But that is not how crediting in software or any other discipline works.

While I don't normally like to kick people while they're down, I am going to make an exception in this case.

I am sick of people thinking their tiny little feature tweak is a startup.

I am also sick of people thinking that computerizing things that people have been doing for tens/hundreds/thousands of years is innovation. It isn't.

If you have an idea for a startup, but your startup could be killed by huge company XYZ if they add one feature, you don't have a startup. If you try to pursue it anyway and finally XYZ Co adds it to their product and crushes you, this should be no surprise. Building a startup in face of such a scenario is an enormous risk, one that'd I'd bet almost never pays off. You should've known better than to waste your time and money.

I have no pity for you.

I am being harsh only in the hopes that more people will learn to respect this hard truth and actually invest their time and brainpower into something more productive and likely to yield returns.

Agreed that "landmark intelligence" as a concept has been known for thousand years but still couldn't be demonstrated technologically till RouteGuru team did it in 2007. Can we figure out the reason?

Can't there be something novel in solving this problem which makes the solution (or perhaps the process) innovative? If it looks like a small feature to you it's your choice.

Adobe holds the patent for Menu based navigation for UIs. That also is just a small feature on a web page. Why does it make sense?

And many a times it's not the concept or technology but the regional strength that's worth taking the biggies head on. Have you ever heard about "Baidu" in China?

Google has still not been able to crack the markets in China/Japan/Korea/Taiwan etc is a big testimony for that.

I urge you to read the post to understand the reason for raising the voice again. The essence is much away from what you have taken it as.

Of course it's possible that it's technically tricky to do. But it still has a small surface area in the UX, making it a "feature" at best of existing mapping systems.

And thinking you can outcompete Google on a "technical" innovation is folly.

I read the ENTIRE post. I always do before commenting.

I guess I am not seeing the same "essence" you are.

I have been doing startups for 15 years and I have seen this type of thing happen so many times. The founders are always shocked when XYZ Co adds the feature and crushes their advantage.

My point is to stress to people that doing something a huge co does but slightly differently is rarely a strong foundation for a startup, and a huge risk to take. People discount the value of the rest of the ecosystem.

Also I used to be more ideological and think that the better mousetrap wins, but frankly it rarely happens.

With whole lot of due respect to your experience, I'd still like to say a few things:

1. If a small feature is tricky to do and has a lot of strategic importance for a bigger platform, it makes complete sense to go for it and also protect the IP for it. We did just that.

You may be aware that Microsoft ended up paying huge money to Ric Richardson for his obvious technology which could have been thought of by 1000s of people before. It's just that he implemented it earlier and protected it as well.

2. >> And thinking you can outcompete Google on a "technical" innovation is folly.

We still feel that our solution is better than Google's even today. It may sound little finicky but we can prove it as well.

3. We were always aware that Google would do it some day. Yahoo did it earlier and we were cool. Microsoft is still struggling to achieve it. We just got hurt because G guys talked to us earlier and their official story looks too hollow.

Let me also confess that we have filed for a patent already and that many people guided us that we can potentially allege Google to be infringing on our patent. But because we still don't have it granted, we're just raising the genuine voice against the pain that RouteGuru hasn't been patronized anywhere despite the fact the Google has admired us for this innovative solution and learnt from us as well.

All fair points.

I am not a big software patents guy, so I don't really like the idea that you can patent this type of thing. Thus I don't like to encourage behaviors whose exit strategy is patent troll, but that's another story.

Your solution might be technically superior, but the less-technically savvy way (ugc or such) is still as useful. Also clearly streetview is designed with the idea of becoming the ultimate landmark-based navigation.

I will yield that I know jack squat about the mapping industry or about what your company does (as that wasn't really what your post was about).

I was just using your post to illustrate a point about something I see happen frequently. It saddens me to see so many people's startups get (predictably) crushed.

While I don't disagree with you, and felt the blog post came across kind of whiny (cultural difference, perhaps), you aren't really responding to his point. He's not sad that he has to compete with Google. He's sad that Google's taking credit for thinking of implementing landmark-based navigation, and is claiming that he or his team met with Google reps a while and that at the time they praised his idea as innovative.

Fair enough, and I didn't talk about the "idea stealing" too much.

But again I have to say that it's pretty obvious thing to do. Even if the GMaps guys were heads-down and no one at Google had thought of it (which I doubt), clearly tens of thousands of people have thought "why can't map directions say turn at gas station X instead of turn at street X".

As someone who lives in an area with lots of missing street signs, I know I thought about that years ago, and I imagine tens of thousands of others have as well.

I've joked about maps and GPSes giving "country directions" for years now. "Turn left where the old store used to be and drive on until you get to where the tornado hit" kind of stuff. It's landmark-based navigation, albeit a jokey one.

In other words, not really a new idea.

I'd call it a stretch to say these guys invented it. There is quite a bit of literature on this technique, e.g. http://research.nokia.com/files/hile_landmark_mum08.pdf

Well, I don't know what this startup is calling his innovation because I didn't tested its system, but what I understand for landmark based is something like:

+Go straight until you see a school at your right.

+Then turn right(left on India).

+Continue until you see a big fountain.

Is that innovation? I don't know in USA but in Spain and a big part of the world that is the normal way of telling someone how to go to a place, is not so strange that another human being has the same idea than you about doing with a computer what you do without it.

That is the main problem about software patents, people want to have a monopoly over ideas a lot of people spontaneously have.

Is that innovation?

The notion of landmark-based direction is not innovative. Using publicly available data sources to cover India in algorithmically discoverable landmarks so that you can give them from two arbitrary points in the nation, that is innovative. (And hard.)

I couldn't even begin to describe a cost-effective way for an Internet startup to discover that my apartment is the second one on the right after you take the left at the statue of Buddah.

This is really eye opening.

1. To learn that some really smart young guys have outsmarted Google so many years ago.

2. How Google articulates stories to make innovation their own.

Had these guys not talked about their disaapointment, how would the world have come to know about the real truth.

Of course it's entirely possible that Google did independently arrive at the same innovation. I have no idea how long Google has been working on this, but it was likely quite some time (which is probably why the solicited help from that company in the first place...)

I'd really be curious to know what the truth is here. Landmark nagivation is an unbelievably big project, and Google is doing it at pretty large scale. It may have very well taken a couple of years to build the thing (not so much the programming, but rather building the data-set itself).

They license the dataset. It is usually the algorithm which determines what is a landmark and what is not (it is my guess).

Stories abound of google talking to innovators and taking their ideas. If you are a startup don't tell them a thing unless their acquiring you depends on it. Until they buy you, they are competitors. Treat them as such.

A detailed comparison of the services by the Routeguru founders themselves

http://www.pluggd.in/routeguru-founders-on-google-landmark-b...

Why did RouteGuru never submit to Hacker News? Did they submit to Reddit, or Digg? How was I supposed to have heard of them?

However, I can feel the pain in a way: a lot of possible projects I can think of are in danger of being made redundant by some Google project or other. Sometimes I worry about my own future - will there even be things to do in the future, without the backing of a huge corporation with ginormous data centers behind it? Should I do everything I can to become part of a giant corporation now, to be on the save side?

But as Yoda says, fear leads to suffering, so probably it is best not to worry about it too much.

Also, I suppose there will always be customers who don't want to use Google out of principle. But Google will probably have the lions share.

And, Google seems to be strange in a way: they do a lot of things, but only a few things properly. Many cool projects they have just end up being incomplete and poorly tended to.

>Why did RouteGuru never submit to Hacker News? Did they submit to Reddit, or Digg? How was I supposed to have heard of them?

Presumably, their target audience is in India, so expecting them to post to sites mostly frequented by Europeans and Americans doesn't make too much sense.

> mostly frequented by Europeans and Americans doesn't make too much sense.

http://www.alexa.com/siteinfo/news.ycombinator.com?p=tgraph&...

Alexa - About 9 % Traffic from India. 2nd to United States.

Collectively, users of those three sites are mostly European and American. And if only 9% of HN users are American, then yes, it does make sense that they didn't target this site.

Correction: "And if only 9% of HN users are Indian..."

I think this idea has been worked on by a lot of people too.

Way back in Hackday India 2007 a group of guys worked on an app called Socialrouting which was hosted at socialrouting.com (it's currently down). It allowed users to mark and view routes with respect to landmarks and also allow people to update traffic in a particular area via an SMS to the SocialRouting twitter account, which was then transferred to the map on the site. I was impressed by the way Pradeep (one of the dev) gave a talk at a Barcamp about their mashup.

And now, I guess that was the idea that morphed into Mapunity http://www.mapunity.in which they are currently working on. It provides live traffic data for a few cities in India. These guys provide services to govt organisations.

P.S: I used their Traffic API in Hackday India 2009 :D thats how I know about them :)

"chase a dream of changing the way directions are consumed in India and globally" - Isn't it a hollow claim ?. Directions were always given in india by landmarks and it's not your firm which has invented it. So trying to take credit for it doesn't sound too good.

In what way did google steal your idea?. It's an idea even kids in devoloping countries know of. and google need not even see your site to realise that the lanamark system is the one that works in this part of the world.

Sory, to sound diffrent. But I don't see this point standing.

No! it's not a hollow claim.

1. We're talking about the technology invention not the concept and RouteGuru pioneered this technology. Also there's a process in which the landmark intelligence is generated which is legible for a patent as well and RouteGuru has filed one.

2. Why Globally? In many countries, landmark intelligence can fundamentally change the way solutions extend directions today i.e. "Street name based directions", which is a very traditional style.

3. I'll encourage you to read the post again. All we felt sad about was that Google did talk to us, learnt whole lot about Landmark intelligent directions from us and eventually it comes up with a lofty story on how did they figure out the problem, terming it as their own invention and without patronizing RouteGuru at all.

We wouldn't have raised our voice if at all RouteGuru could get a single mention anywhere.

All we felt sad about was that Google did talk to us, learnt whole lot about Landmark intelligent directions from us and eventually it comes up with a lofty story on how did they figure out the problem, terming it as their own invention and without patronizing RouteGuru at all

Can you prove that? How do you know that they didn't truly come up with the idea on their own?

There are thousands of possible ways this thing played out. Maybe the discussions you had were after some people at Google started having that idea. Maybe the people you met had never had that idea, but another employee in the next cubicle was already working on it for months.

You don't know how it all worked out. Parallel inventions are common place, especially in this case, since it seems that the idea was fairly obvious and in widespread use in India.

I think what they are angry about is that Google claims to be the first one to do that (when their startup has been doing this for long) and since Google is so well knows the world is thinking they made a local (India specific) innovation.

Yahoo! India maps version has been having it for over a year or two now, which I have used myself few times.

That's a great point. Why didn't RouteGuru raised a voice against Yahoo? Let me explain:

1. Yahoo din't talk to RouteGuru team before the launch. 2. Yahoo also din't come up with any false story either and there team did mention about RouteGuru in one of their Blogs.

I urge you guys to understand the point of disappointment well.

Reply on news.ycombinator.com