Slashdot Mirror

← Back to Stories (view on slashdot.org)

OSGeo Foundation Up In Arms Over ESRI LAS Lock-In Plans

Posted by timothy on from the now-it's-ours dept.

Bismillah writes: The Open Source Geospatial Foundation is outraged over mapping giant ESRI's latest move which entails vendor lock-in for light detection and ranging (LiDAR) data through its proprietary Optimised LAS format. ESRI is the dominant company in the geospatial data arena, with its ArcGIS mapping platform boasting with over a million users and 350,000 customers.

35 comments

  1. Their software cost an arm and both legs yet... by Drakker · · Score: 5, Interesting

    I'm not surprised with this. GIS is almost a mono-culture that has been dominated by ESRI since forever. Their software costs in the thousands, yet crashes all the time and a lot of the included tools just don't work. Some parts of the interface have not changed since the 90s and they keep building on this dysfunctional foundation. Working with ArcGIS is a pain in the rear, yet for a lot of what ESRI software does, there is no alternative. Whenever I can I code my own stuff (using GDAL http://www.gdal.org/) and do all I can in QGIS (http://www.qgis.org/), but for a lot of tasks, you are stuck with ArcGIS and other ESRI tools. The market is more than ready for a new player that will make reliable software (whether commercial or open source, doesn't matter to most as they are used to pay through the roof for ESRI software anyway).

    1. Re:Their software cost an arm and both legs yet... by Richard_at_work · · Score: 1

      Do ESRI actually generate the data? If so, what's the complaint?

    2. Re:Their software cost an arm and both legs yet... by TechyImmigrant · · Score: 2

      Do ESRI actually generate the data? If so, what's the complaint?

      Nature generates the data. ESRI just hoard it.

      --
      I should use this sig to advertise my book ISBN-13 : 978-1501515132.
    3. Re:Their software cost an arm and both legs yet... by hey! · · Score: 3, Insightful

      Don't forget training and consulting. That brings in quite a bit of revenue too. In my experience most ArcGIS installations aren't actually functional because they don't have people who can work the software. Even after training most installations don't have the personnel to dedicate to keep up with it; they maybe produce a report or two, and then the software sits on the shelf, then they need to send someone else to training.

      In this environment a lot of people using ArcGIS might as well be using QGIS. If there were training and support for QGIS, this would build a user base which would attract developers. I think a lot more could be done in getting users to adopt web based mapping -- WMS and WMF -- too. The web is such moving platform that the kind of desktop based entrenchment ArcGIS enjoys is less significant.

      --
      Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
    4. Re:Their software cost an arm and both legs yet... by Anonymous Coward · · Score: 0

      There's always Intergraph/Hexagon prodcuts, but they don't have the market share that ESRI does (think interoperability) and their licenses are expensive as hell too.

    5. Re:Their software cost an arm and both legs yet... by nava68 · · Score: 3, Informative

      No, but as soon as the Market leader does propagate a new data exchange format, the actual creators of the data (commercial companies as well as state agencies) will follow. ESRIs market share is dominating and unfortunately many GIS users are not interested in open tools. So if a more efficient and comfortable format will be available then the GIS community will willingly accept it - without even considering the problems of a further lock in. In addition to that, the Open Source Geo community is already providing the LASzip format for the same functionality and that format is far superior to the classic one and probably as efficient as the one ESRI is developing - plus it is open, documented and all, albeit so far not subjected to the process of standardization.

    6. Re:Their software cost an arm and both legs yet... by Urban+Nightmare · · Score: 3, Interesting

      I work for a Government Agency that uses ArcGIS for that last 10 or so years. Personally I'm a Linux user and try to use and contribute to open source projects (although lately I haven't been able to). I also use QGIS but find it's not always what I want when it comes to the cartography part of GIS. The maps it produces never seem to be as nice as ArcGIS. Sorry got a little off topic here. What I'm saying is that even though I like and use open source, my bosses don't give a rats ass about it. If you even mention no licensing fees then they think its an inferior product. They have drank the kool-aid, so we pay tens of thousands for licensing of all kinds of products that have the same or better open source alternatives.

      I believe this is also how many other agency view things. They don't care about open source. They just want someone to SELL them what they need.

    7. Re:Their software cost an arm and both legs yet... by Anonymous Coward · · Score: 0

      Some parts of the interface have not changed since the 90s

      How horrific! I'm glad open source gives me the choice to use Firefox in Gnome 3 at least.

    8. Re:Their software cost an arm and both legs yet... by Drakker · · Score: 2

      I get what you mean, but these parts of the interface are extremely clunky. But even worse than that, they migrated all their tools to an ActiveX interface that relies on Internet Explorer... and it really isn't all that stable, randomly stopping to work (you have to restart ArcGIS when it happens).

      But even worse than that, it requires IE to be set at the default security level, which is not all that secure. Since I don't use IE, and want to minimize the risks to my lab machine, I usually set it to the maximum security level everywhere, which in effects disables big parts of it. While not as ideal as completely removing the software would be (if it was possible...), it reduces risks considerably. But wait a second... you guessed right. ArcGIS will just not allow you to use any of its built-in tools when you do that. How wonderful. Who's the genius that thought it was a good idea to have such a large piece of software being dependent on the worse web browser ever in existence?

      So while I agree with you that "updating" interfaces can go too far (yucky gnome 3 is a good example), in this case, an update is badly needed. But I suppose they will be forced to do it soon enough, with MS supposedly discontinuing IE and touting their new browser that is possibly not just a fresh coat of paint...

    9. Re:Their software cost an arm and both legs yet... by Anonymous Coward · · Score: 1

      >>>Working with ArcGIS is a pain in the rear

      Wholeheartedly agree. I am one of those poor souls who has to use this tool on a regular basis in my job and it is almost infuriating at every turn. Believe me the bugs that I encounter are sort of incredible at this day and age of software: for instance: there are specific addresses when fed to the ESRI geocoder service will actually crash the entire software! And their map rendering sucks donkey balls. The fact that there has been no challenges to their dominance is very telling of the defeat of free market at the hands of crony capitalism.

    10. Re:Their software cost an arm and both legs yet... by discord5 · · Score: 1

      I'm not surprised with this. GIS is almost a mono-culture that has been dominated by ESRI since forever

      The attitude is slowly changing though, with more and more open source tools becoming available. Like you said, gdal and qgis are adequate for most common use, but in the case of LiDAR there's PCL (point cloud library) offering a workable alternative for the traditional las/laz tools in some cases. On the server end there's PostGIS which is a really nice set of geospatial procedures for Postgresql which for a lot of uses is more than enough. There's geoserver, mapserver, and the various html/js frameworks like openlayers and leafletjs. If you're a bit on the creative side and have the expertise (or moderate experience and willingness to learn) in the field of remote sensing you can build quite a few nice applications.

      The market is more than ready for a new player that will make reliable software (whether commercial or open source, doesn't matter to most as they are used to pay through the roof for ESRI software anyway).

      There's been a few older smaller players that I've been in contact with that are not so happy with the open source tools available as they're seeing competitors pop up getting a headstart in their dev cycle with lower initial costs. The barrier of entry has been lowered really, which is a good thing, because a lot of the smaller players were feeling too comfortable in their own niche applications as well. Personally, I take great pleasure in watching those new companies come along and use and improve these open source tools and try to upset what is traditionally a very "embedded" market.

      Sure, ESRI will still dominate the market for some time to come, but I think that they realize that in the not-so-far future they might lose that position. Moves like this fileformat seem like an indicator to me, and I wouldn't be surprised if they start pushing their format on instrument builders or the people involved in large scale data acquisition. They wouldn't be the first to do this, since las/laz by itself is not the best of formats for quickly seeking through unless you start building indexes (and even then it can still be ... cumbersome). I've talked to plenty of people who convert their flightlines into nice quad- or octrees for their own tools with a dash of compression on top. Personally I'll still be asking for either las/laz or XYZ data. To me, the data itself is just a means to an end really, so I'd rather have a format I don't have bend over backwards to get what I want.

      As for the price issue, for most people dealing with large enough datasets the real cost is in data acquisition, flight planning (when airborne), storage, and if you work in countries with strict privacy guidelines wrt remote sensing data there's a lot of cost in managing your data according to certain policies and oversight. If anything, that last one alone can be quite migraine inducing, although it's less the case with lidar I guess..

      But I am noticing that in more and more environments where there's research being done, or where there's active development, people are more and more embracing other tools, often open source, which gives me a warm and fuzzy feeling inside. I've been seeing a lot of pre- and post-processing in python with gdal and numpy lately, and you notice on the internet that there's more and more people picking it up.

    11. Re:Their software cost an arm and both legs yet... by Anonymous Coward · · Score: 0

      "Hoard" explains it perfectly.

    12. Re:Their software cost an arm and both legs yet... by dwywit · · Score: 1

      Sounds very similar to my own situation 10 years ago.

      ArcGIS, while expensive, produced useful output, even to the point of other councils licencing our work.

      Along comes a new IT Manager (my direct superior). Wanted to know why we were using BSD and not Windows for our internet-facing servers. Seriously, his question was "What's BSD?". He didn't trust anything without a gui, and was deeply suspicious of the Unix server running ArcGIS.

      --
      They sentenced me to twenty years of boredom
    13. Re:Their software cost an arm and both legs yet... by Xyrus · · Score: 1

      ESRI is the Tomacco of GIS software. It's terrible, but they can't stop using it.

      I was pulled onto a legacy application that was using ESRI, and after seeing what a disaster it was (the project was on the verge of losing funding) I trashed the whole thing and rebuilt it from the ground up using open source tools and libraries. What used to take weeks to months of hair pulling frustration from programmers now takes a couple minutes for a non-techie user.

      Anecdotes are a dime a dozen though. I'm sure ArcGIS and friends have their uses, and in some cases may be the only game in town for some features. I just haven't found where'd I use ESRI vs. an open source equivalent.

      --
      ~X~
    14. Re:Their software cost an arm and both legs yet... by c_g_hills · · Score: 1

      For such a large software company, their bug tracking is a joke. I reported an issue where the select by location tool was not selecting accurately, and in the end they admitted it was a known bug with high severity but no planned fix date. Now this makes me anxious because I am worried that there could be other bugs affecting the accuracy of my work, but the bug tracker does not let you see all the open bugs for a given product, nor can you subscribe to updates on existing bugs!

  2. CJMTK - ESRI lock-in, mandated by Congress? by david.emery · · Score: 2

    OK, this is hearsay, but I remember being told maybe 15 years ago that Congress placed language in the DoD budget mandating the use of a Commercial Mapping Toolkit, with language such that only ESRI's product would qualify. What I know is the CJMTK (Commercial Joint Mapping ToolKit) is an ESRI proprietary interface that is mandated for use in DoD systems.

    Since then, it's been difficult to provide alternatives, particularly at the library/component level. Google Earth has gotten some traction, but not as an API but rather as a rendering engine.

    1. Re:CJMTK - ESRI lock-in, mandated by Congress? by jandrese · · Score: 1

      I see language all the time in Defense related project that require all software to be from "commercial vendors". It's one of the big reasons Red Hat Linux stays in business. There are reasons for it. If something goes wrong and there needs to be someone to hold accountable you can't just call up random screen names from GIThub and hope they appear before Congress.

      --

      I read the internet for the articles.
    2. Re:CJMTK - ESRI lock-in, mandated by Congress? by david.emery · · Score: 1

      There's a difference, though, between a preference for COTS with vendor support, and a mandate for a specific COTS product. An Open Source product without anyone providing maintenance is a risk. An Open Source product where you can -compete- for maintenance is a real benefit. A COTS product where you pay whatever the vendor charges for maintenance is at least a predictable life-cycle cost, but that vendor has you by the short-hairs. I've seen products where the sustainment contract was a lot more than the purchase/license price, and there was nothing you could do about it.

    3. Re:CJMTK - ESRI lock-in, mandated by Congress? by Anonymous Coward · · Score: 0

      Was CGI Federal called to appear before Congress? The concept of "accountability" for software is the biggest red herring in the proprietary vs. free software debate. Microsoft's never going to be accountable for software failures, whether to your mom & pop business, a multinational corporation, or the DoD itself.

    4. Re:CJMTK - ESRI lock-in, mandated by Congress? by Anonymous Coward · · Score: 0

      That sort of thing is the the defense procurement equivalent of hitting a grand slam.

      The savvy salesman/bizdev fellow works very hard to get the customer to write the RFP such that only one product, their product, qualifies. Then the contract can be put put to bid safely, knowing that the competition is moot. The product doesn't have to work and can be expensive as hell, but it is the only one that can meet criteria X placed there specifically to exclude all other. Actually getting the poison clause written into legislation means that you can just coast as long as you keep the decision makers happy/

    5. Re:CJMTK - ESRI lock-in, mandated by Congress? by Anonymous Coward · · Score: 0

      The current DOD thinking is not to be stuck in proprietary stovepipes for software. They are pushing using OGC ( http://www.opengeospatial.org/) standards for web services and file formats. They are very interested, in particular, with the new OGC geopackage format , http://www.geopackage.org/ ( a specific format of sqlite database) as an alternative to proprietary file formats for geospatial data exchange.

         

  3. Light detection and ranging?? by friesofdoom · · Score: 1

    From http://en.wikipedia.org/wiki/L...: Although thought by some to be an acronym of Light Detection And Ranging, the term lidar was actually created as a portmanteau of "light" and "radar".

    Go fuck yourself.

    1. Re:Light detection and ranging?? by Fortran+IV · · Score: 2

      ...And the DAR portion of RADAR stood for "Detection And Ranging". So why the nerd rage? Someone just chose to deconstruct a portmanteau into its root elements.

      --
      I figure by 2030 or so my 6-digit UID will be something to brag about.
    2. Re:Light detection and ranging?? by DRJlaw · · Score: 2

      Although thought by some to be an acronym of Light Detection And Ranging, the term lidar was actually created as a portmanteau of "light" and "radar."

      Go fuck yourself.

      Although expressly used by many as an acronym of Light Detection And Ranging, the first usage of a term magically enshrines that usage, and no other, as the exclusive meaning of that series of letters for all time.

      Or not.
      Portmanteau, noun, plural portmanteaus, portmanteaux [pawrt-man-tohz, -toh, pohrt-, pawrt-man-tohz, -toh, pohrt-] (Show IPA). Chiefly British
      1. a case or bag to carry clothing in while traveling, especially a leather trunk or suitcase that opens into two halves.
      2. Also called portmanteau word. Linguistics, blend (def 10).

      Lidar was certainly not created as a case or bag of "light" and "radar," and your intended reference to a second meaning is obviously irrelevant because FIRST!

  4. SDTS Anyone? by Anonymous Coward · · Score: 0

    Spatial Data Transfer Standard. It was a good idea that was extensible and steamrollered into oblivion by, um, er, market forces. Hey if they own the data they can package it however they want. If not, well that is what open standards are for.

  5. Re:Illiterate American fucks by Anonymous Coward · · Score: 0

    You did see that the poster was from New Zealand? Any fool could have checked that in half a second.

  6. Nothing new from the ESRI'alites... by dewright_ca · · Score: 2

    This really isn't a new event; just as always with ESRI you will see them partner with business to define a need and then extend those functions to be new tools in their product only to see a short time later a ESRI specific toolset that is given away for free that does what the partner did but locking you into there platform. But in this they drive business away from said partner.

    Having been on this side its hard on a small business to not see them for the predatory company they can be. So many of the "NEW" things I have seen from them in the last 6-7 years are not new but they can spin it to make the followers of the great Redlands Prophets just write blank checks...

    --
    He who is always at the bottom of the distribution list, but needs the information first!
  7. Not quite a monoculture by grilled-cheese · · Score: 1

    ESRI does have some competition, just not a lot. You can find GIS shops that run Autodesk Map 3D (merged with autodesk topobase a while back) and Intergraph. Which of the 3 you use is largely dependent on your region. Intergraph is fairly popular outside the US. Autodesk tends to be more used in the western US, and ESRI is about everywhere else.

  8. PostGis (SQL) + Python + Web = No GIS needed by Anonymous Coward · · Score: 0

    Treat data like data, put it in a database. Analyze, develop patterns, put into action. Repeat.

    ESRI propagates bad practices and patterns.

  9. History Repeats... again. by Anonymous Coward · · Score: 0

    But the announcement was met with outrage from the open source geospatial community, which claims by creating a proprietary variant of the LAS format instead of improving on the available open formats, Esri was negatively affecting software interoperability and creating format fragmentation and vendor dependance.

    "The Optimised LAS format is neither published, nor available under any open license, which provides both technical as well as legal barriers for other applications reading and/or writing to this proprietary format," the letter stated.

    This same problem has happened with every open container format that's ever existed. AIFF, TIFF, AVI, MPEG, MKV ad nauseum. It's all well and good having an open container format but the moment you allow freeform CODEC formats within it there will be corporate assholes that want to subvert that popular container format for their own gain.

  10. There's already another LAS format by Anonymous Coward · · Score: 0

    "It also requested that the term "LAS" not be used in the name of a proprietary format, as it has the potential to be confused as an approved associate of LAS."

    Which is a bit comical because LAS format was already a term in use for an entirely different data format for storing well geophysical data since the 1990s.

    It doesn't excuse ESRI's nonsense, but the open format advocates probably should have picked a different acronym for a LiDAR data format too. I swear, do people even LOOK to see if an acronym and format name is already in use before going with it?