Open-Source IT for recreational flying

[Jan Olieslagers]

I have done much pondering and some little work on creating my own I/T for flying. As it stands now, ir runs on an Asus netbook and produces a moving map with aerodromes and navaids displayed, along with planned and effective routes. Limited as this seems, it has once allowed me to find back my field when I had unwisely left the circuit in poor visibility.

The more I proddle about, the more obvious it becomes I can never turn out anything really useful on my poor little own self. So I thought I'd launch a discussion thread, to find out if others are thinking and/or working along the same lines, and see if perhaps a TEAM could be set up.

Resources:
-) a database with all kinds of info (a/d, navaids, airspace, roadmap, extended airfield info, &c &c)
-) GPS
-) some source of real-time Wx that is available up there?
-) data of other traffic either from 1090-ES receiver and/or flarm-receiver and/or ads-b
-) flight plans either to be created on the machine or imported (as gpx or whatever)

Pilot info supplied:
-) moving map with display of planned and effective routes
-) optional Wx overlay
-) terrain overlay
-) separate display (text mode?) of various data (ground speed, gps-altitude, gps-time, ...)
-) collision avoidance based on available info
-) early ground warning if elevation data available

Hardware:
-) some kind of portable PC, netbook or whatever
powered from the plane but with batteries for at least 30 minutes in case of electrical failure
-) gps-receiver
-) optional receivers for 1090-ES, ADS-B, flarm to allow traffic avoidance

Software:
-) linux seems the obvious choice of O/S
-) application(s) could be developed in whatever language, the most obvious seem C, C++, Java ; availability of libraries would be an important factor

Advantages:
-) integration would reduce the total number of gizmo's
-) open source would mean vendor indepence
-) extremely affordable, except probably for a sunlight-readable display of practical size (7 inches at least?)

Disadvantages:
-) neither the database nor the "other traffic" could ever be guaranteed to be complete
-) keeping the database up to date would be a huge and enduring task, requiring a separate well-coordinated team
-) psychologically, such a system would be dangerous because it would SEEM to supply very complete information, leading pilots to depend on it blindly

Ideas and comments welcome!

[IO540]

The Q is what exactly you are trying to achieve.

Pre-flight, the job is well defined: route planning, notams, weather, airport contacting for PPR/PNR/etc.

This is easily done with a very common item known as a "laptop" :) You can get them from £100 to £2000, and a normal windoze one will run everything needed including - at the top of the GA food chain - Jeppesen stuff.

Airborne, people look for different things. A VFR pilot might want to run a decent GPS moving map (something better than you can buy, or something which can run downloaded/"shared" maps ). An IFR pilot might wanto to see approach plates. Both might want to get weather data, if it is a long flight.

The platform for this is less clear, largely due to a near total lack of sunlight readable hardware which is light, thin, etc. A lot of people are looking at solutions but most of them deliver only a part of the job, so one would end up with a pile of gadgets, which need charging, etc.

[Justiciar]

At the risk of just pouring cold water on the idea, hasen't this already been done?

There are a number of pieces of flight planning software available for windows which already do this: NavboxProPlan, very well established with an annual subscription of < £50 and Skydemon, which is more expensive but has better graphics. Both can be used as a moving map with suitable hardware. Skydemon also has a mobile version for PDAs. Jeppesen (which I have no experience of) also provides flight planning (not sure about moving map). You then have Flymap, Airbox and MemoryMap which offer CAA or Jeppesen charts, with the first two also having PC based flight planning software. These all run on small units with Memory Map being available for PDAs and the iPad. Then of course you have Garmin, Bendix King etc touch screen devices.

Last week end I had the chance to play with a friend's iPad. Unfortunately it was not the one with built in GPS but I have to say that it seems an ideal cockpit companion. You can get a screen protector which I believe improves the readability in sunlight. I am investigating the non GPS/3G one but using bluetooth to connect to a separate GPS module or a mobile phone for data download when required - a far cheaper option than the built in GPS/3G models. The unit has long battery life, in contrast to the units mentioned which generally struggle to last beyond about 90 minutes.

The big issue is that there is little in the way of aviation software for europe as opposed to the US for the iPad. This seems to me to be a worthy software challenge; I know that the writers of the programs mentioned above have no plans to do an iPad port of their programs.

[Jan Olieslagers]

OK, seems I wasn't clear enough. There is indeed plenty of stuff available for planning - I am talking in-flight. Wasn't that obvious when I mentioned collision avoidance and early ground warning? IO540, you had better read all of the text before smiling about laptops.

The idea is indeed to have as much functionality as possible in as few boxes as possible. As mentioned, finding a suitable yet affordable display is a major hurdle.

And it is a basic requirement that all software and data should be freely available - which precludes Windows. The price for using free data is that it could never be certified.

@Justiciar: what difference between software for the USA vs. Europe? Temp's in °F and that kind of fun, but basically I can see no differences. Again: getting the data in a useable format, and keeping the data up to date is the real challenge.

[IO540]

all software and data should be freely available - which precludes Windows. and making it understandable only to a tiny minority.

And I say this as a software/hardware developer, having programmed in assembler and C on 8048, 8051, z8, z80, z180, z280, 80x86, Atmel 90S, h8/300, h8/500, various FPGAs, ASIC design, CP/M2.2, DOS, windoze, Fortran, Pascal, Basic, gawd knows what else... I think my anorak qualifications are impeccable, despite no longer being a paid up member of the British Subterranean Society :)

The price for using free data is that it could never be certifiedand it will also be full of errors, once away from the well trodden GA routes. Fortunately data doesn't have to be certified if used in a portable device.

But this is why people pay money for flight planning software. Somebody is sitting there, getting paid for populating the database.

Anybody can write a flight planning program which totally blows everything currently out out of the water (well, certainly this was true when Navbox or Flitestar were the only games in town). It should take a few months, max.

Getting the data is the huge challenge. It is easy to rip it off from certain well known commercial products but then you cannot openly supply/sell the result. Even at the IFR level, where things should be easy, you can get the airway database but you cannot get the SIDs/STARs...

[Justiciar]

You did say:

flight plans either to be created on the machine or imported (as gpx or whatever)

... which implies some ground based functionality.

what difference between software for the USA vs. Europe? Temp's in °F and that kind of fun

I was talking about the iPad, and I would have thought that maps, airfield and navaid data, airspace data, ability to generate plogs, Rnav data, frequencies is what is required, so a little more than temperatures in F and pressure in inches!

keeping the data up to date

But as I said, several manufacturers/producers of software are already doing this. Will people be willing to pay a modest fee for an established product which hopefully guarantees the accuracy of the data or pay nothing but use date of questionable accuracy or provenance? Open source software is one thing, open source data is another! I would question whether reliance on such data enables a pilot to comply with his duty to self brief using reliable data.

It seems to me you highlighted some fairly insurmountable obstacles yourself in your first post.

[stevelup]

I'm moderately anti-Windows - I think it is rarely, if ever the most optimal solution for any application. All my commercial work is undertaken using Linux, and all my personal fixed and portable devices are Apple.

However, there is no point in suffering and re-inventing the wheel just for the sake of it.

Sky Demon does a good chunk of what you are asking for and runs as both a desktop application and on a Windows CE portable. The developer is highly responsive and adds new features once a few people have expressed a desire for them.

So my hand was forced...!

I've got a netbook for planning and a cheap 7" Chinese Windows CE 6 GPS for use in flight. Screen brightness is a problem but it seems just about tolerable - I'm looking at I fashioning some kind of hood for it.

[Jan Olieslagers]

Yes, I did mention some ground functionality as well, OK.

I'm unclear about the "insurmountable" though. I am wondering
-) how good is the actual "paid" data - ISTR several flaws mentioned. Even if the "free" data is not perfect,
-) how hard people would find it to verify the relevant data before their flight.

After all, even when using Jeppesen or whoever as a data source, isn't one supposed to cross-check with the relevant AIP?

Then again, drawing a parallel with streetmaps: when I compare the "free" OpenStreetMap with commercial maps widely available on the internet, I find the free data generally more elaborate and more up to date. I certainly hope to include data from the OpenStreetMap database in my project one day, such as major roads, railways, large surfaces of water. If a sufficient team couild be formed over the years, the database has every chance to become just as complete and as dependable as its commercial counterparts.

[IO540]

-) how hard people would find it to verify the relevant data before their flight.

After all, even when using Jeppesen or whoever as a data source, isn't one supposed to cross-check with the relevant AIP?

None of that is feasible, at all. In planning and in flight, you rely 100% on the databases. Almost nobody ever reads any AIP - except occassionally for airport details like whether it needs Customs PNR.

Then again, drawing a parallel with streetmaps: when I compare the "free" OpenStreetMap with commercial maps widely available on the internet, I find the free data generally more elaborate and more up to date

Probably because they ripped it off high quality (by which I do not mean TomTom which uses databases several years old) commercial products. You can do this if the end result is thinly spread (e.g. open source data) or if you sell it but keep a low profile.

Due to the widespread disgust over tight copyright control of State maps (most of them having been produced at taxpayers' expense) there is a massive amount of underground map distribution. Do a google on Kabouterbond for one of several "movements". Not much aviation data there but all the stuff can be found on torrents anyway, and current too.

This is why a windoze tablet is a good solution for in-flight moving map GPS. You get Oziexplorer and you can get Ozi maps for all of Europe, and ONC maps for most of the rest. If there was an Oziexplorer for the Ipad, that would interest a lot of people, but one could say the same about Jeppview, Flitestar, Navbox, Skydemon, etc.

[Jan Olieslagers]

Then again, drawing a parallel with streetmaps: when I compare the "free" OpenStreetMap with commercial maps widely available on the internet, I find the free data generally more elaborate and more up to date Probably because they ripped it off high quality (...) commercial products.

That's a grave accusation. On what factual information is it based? If you have none, it comes close to slander.
I know for a fact that at least SOME of the OSM data is hand-crofted, actually I contributed to several mapping parties to help build up data for my country - which is regrettably incomplete, at street level.
And the project certainly makes every effort to remain on the legal side, and to discourage ripping wherever they can.

[IO540]

OK, how does one compile street maps?

Method 1: you go around and survey the town, and draw your own map.

Method 2: you use an existing map.

In aviation, the official method is to lift it from the AIPs. The data is not directly machine-readable though, so most products get it from other sources. Pre-2006, it used to be lifted from the U.S. DAFIF but that is now closed and most low cost U.S. products shut off their European coverage when that happened. Today, you would do a combination of things e.g. a B2B deal with Eurocontrol, or maybe running off various PDF reports from EAD and parsing them, etc.

But if you get successful, lifting AIP data will draw attention. Jepp got sued for this in Australia, and did some kind of confidential deal with them, and probably with other AIP publishers who smelt money after the Australians won.

[Jan Olieslagers]

<<off-topic>>
OK, how does one compile street maps?
Method 1: you go around and survey the town, and draw your own map.
That's how'tis done, in a team, at mapping parties. It looks rather funny, sometimes downright suspect, while gathering the GPS tracks. Afterwards, when the tracks are uploaded and streets are drawn from them, it is very nice to see the streets coming together bit by bit.
In this last step, existing maps may be used as a secondary source of information; but I have on at least one occasion introduced bad data by blindly believing a commercial internet map provider. Better secondary data comes from one's photography during the gathering, and from official maps, e.g. at the municipality.

[Jan Olieslagers]

lifting AIP data will draw attention

How so? Isn't the data in the AIP public, by its very nature? I can understand that the officials don't like commercial use of their efforts, but how could they legally object to non-profit distribution?

[IO540]

That's how'tis done, in a team, at mapping parties. It looks rather funny, sometimes downright suspect, while gathering the GPS tracks

Clearly, I must renew my British Subterranean Society (http://www.subbrit.org.uk/rsg/roc/index2.shtml) membership ASAP :)

Isn't the data in the AIP public, by its very nature? I can understand that the officials don't like commercial use of their efforts, but how could they legally object to non-profit distribution?

You could ask Jeppesen what deal they did. Nobody I know knows...

I don't suppose they can object to nonprofit distribution. But the nonprofit concept tends to break down when it comes to more specialised areas like aviation. Too many potential users but too few people willing/able to put in the huge effort.

[Justiciar]

The fact that data is in the public domain does not mean that one can make unrestricted us of it. Given that the purpose behind publication is to comply with an international treaty obligation under ICAO the NAAs would be quite entitled to place legal restrictions on its use for other purposes. The copyright in the AIP remains with the NAA (i.e. the CAA) and they can license it however they wish. Ditto with anything else put into the public domain, unless the owner makes it clear that unrestricted use is allowed.

I have no doubt that incorporating data from the AIP into a piece of software, whether open source or otherwise, would be a breach of copyright and restrained accordingly.

[Gertrude the Wombat]

And it is a basic requirement that all software and data should be freely available
That might be your requirement as a developer, but it's difficult to imagine it as an end user requirement.

The end user wants

- reliability of software, and support with problems
- reliability of data, and someone to pass legal responsibility on to if the data is wrong (eg resulting in a RA(T) bust)

and stuff like that. They are unlikely to care very much whether the developer is a member of the open source religion, and most will not mind paying the equivalent of a few minutes' flying for something that meets the above needs.

[NazgulAir]

Have you looked at PocketFMS? Features (http://www.pocketfms.com/2-Features.asp)
- multiple platforms
- multiple interfaces
- usable on the ground and in flight
- up-to-date map data, terrain, AIP data, MET data
- low cost subscription
I use it only on the ground for flight planning, but I've seen it used in the air.
The maps are very good. They use their own maps with overlays. There's a ton of features.

[Rod1]

Just join the existing group doing what you want;

Home of the PocketFMS Foundation. (http://www.pocketfms.com/)

Rod1

[Jan Olieslagers]

PocketFMS is NOT doing what I want. I'll try to remain polite but have you actually read my message?

[Jan Olieslagers]

someone to pass legal responsibility on to if the data is wrong (eg resulting in a RA(T) bust)That is a crucial point, Gertrude. The system I envision certainly could not offer this, neither does it want to try. But IS any kind of guarantee offered by commercial airnav data suppliers like Jeppesen? Are there any stories known of people getting compensation for errors due to incorrect commercial data?
IOW, what criteria must be met by suppliers of airnav data for pilots to rely on them as a "reliable source"? Are companies like Jeppesen (or their products) checked by any authority?

For myself I am stupefied to see almost everybody relying blindly on commercial stuff. Apparently very few people consult the AIP, although AFAIK it is the only official source of information. Up to now I have never relied on commercial data: I planned my flights - up till now very limited both in number and in area - on freely available data, than checked and confirmed on the AIP and official maps.

[Justiciar]

We've all, I suspect, read your message, including

So I thought I'd launch a discussion thread, to find out if others are thinking and/or

So you've had a lot of what others think. I detect a marked prickliness at views expressed which do not appear to reinforce your thoughts.

Most if not all the features you highlight are available in other products, including traffic avoidance on PocketFMS (I don't know about weather downloads - I understood that this was not available in the UK/Europe even to panel mounted units, but I will be corrected). True they are not free to users but those that are available have established a reputation for reliability and data integrity and have an established user base of faithful followers. Given the relative tightness of the market and a number of established products I cannot imaging this being viable. Having a product which will be dependant on regular data updates means having a robust business model with stable finances in place to ensure that users receive consistent and accurate updates. I don't see this being done by a team of what would essentially be volunteers.

[windriver]

if others are thinking and/or working along the same lines, and see if perhaps a TEAM could be set up

I'm always reluctant to appear negative when someone comes up with an idea and asks for comments - but the system you describe (unless you hit it lucky with a commercial backer) would most likely have to come with a whopping great health warning - Not To Be Used For Air Navigation" (for reasons already stated)

As a software person you will have asked yourself the question many times - "Am I proposing to build an application to solve a problem?" - Or building an application that's looking for a problem to solve...

Whils't it looks like the project looks "relatively" straightforward from a technical perspective the data sourcing and updating, system updates, and all the other baggage that inevitably accompanies any software release looks like it might be extremely time consuming even for a small army of volunteer coders and helpers.

Much of the best open source stuff started life in a university or as a business application with people being paid or funded to do all the graft..

Prove to me there is a 'market' for this and I may be interested in doing a bit for "the team."

[Jan Olieslagers]

@Justiciar: I do know my own tendency to "pricklyness" - what a neat word! - but had tried to limit it to people not reading what I actually wrote. I have learned to tolerate people disagreeing... };-) especially after asking myself...

As for "products" being available: I am fully aware of several, PocketFMS not the least - but they all remain products indeed. I revolt at the idea of having to pay for information that is _BY_LAW_ and _BY_NATURE_ freely available. And indeed what I have set up as yet is just a simple but usable graphical representation of data I collected, mostly from the AIP's, or from other sources who themselves consulted the AIP's.
Commercial sources carry my continued distrust - money gone is a certainty, the quality of supplied data remains to be seen. Hence my curiosity about liability/warranty. The lack of response is not really surprising me.
DAFIF I came to distrust, probably undeserved, through its advertising the wrong tower frequency for my (then) home airfield for several years. And it is discontinued, anyway.

The effort required to keep up the database would be big, but could be largely automated. Keeping users' data up to date would surely be their own responsability - the project can only endeavour to keep its database to the best possible standards, and offer a tool for users to synchronise with it.

@Windriver: thanks for your constructive reply. Yes I do know the matter of solution vs. problem. No I do believe I avoid this pitfall, I am trying to set up some free software and free data sources - "free" meaning both free of cost and free of copyright or other limitations - to do as many useful things as possible through the same hardware, to limit both cost and complexity and power consumption and panel occupation.

One "subproject" that would have to be done anyway is to create a European superset of aerodrome information, going beyond the AIP information (such as availability of fuel, restaurant, taxi, overnight hangarage). Several countries have comprehensive systems, but I find a European consolidation sadly missing. A multi-lingual consultation interface should be the least trouble... Again, such a database could be automatically updated from the various national contributors.

[IO540]

But IS any kind of guarantee offered by commercial airnav data suppliers like Jeppesen? Are there any stories known of people getting compensation for errors due to incorrect commercial data?

Not as such but if you flew with say Jepps and did a bust, you can say to the CAA "look here is my chart, it is current" and if there was actually a mistake on it (which is hardly unknown with Jepp ones) they would find it very hard to prosecute.

There is almost no insurance angle here because flying with hopeless charts is negligence, and insurance covers negligence.

There is no law (USA or UK) which stipulates a particular chart must be carried or used. It comes down to due diligence by the pilot.

I would guess Jepp have product liability insurance so if one of their approach plates is wrong and somebody gets killed, they are insured. I vaguely recall this has happened somewhere but it must be extremely rare.

But there is almost zero risk of anybody claiming over a mistake on a VFR chart or a "VFR" flight planning program because VFR=VMC and only a d1ck will fly into a hill in VMC ;) IMHO, IANAL, if you sell a "VFR" flight planning product then provided you make sure the mapping data supplied is obviously too poor to use alone (e.g. Navbox) there is no chance of ever getting successfully sued.

[Jan Olieslagers]

Rather than flying into a hill - a very limited risk to anyone in this flat country - I was thinking of the real case of a VFR map published by our government map printers which gave incorrect vertical limits for some TMA. I never heard of any subsequent litigation, but it would certainly have been interesting.

[IO540]

A prosecution by the CAA would be a criminal prosecution, but they would not be able to do it if it happened due to an error on a current chart of a type generally used for the purpose.

Normally, "getting sued" refers to a civil liability and I can't see how that would materialise. I suppose if you busted CAS and brought down an airliner, then the insurance co. which paid out to the victims could go after the chart publisher, but this is such a remote chance.

[Rod1]

Jan Olieslagers

I read what you put. I also read the history of PFMS, which started out with what I thought was exactly the same intention you have. PFMS was initially FOC, then on a donation, and then on subscription. It appears to do all the things you want. But then you know all that and did not need to ask in the first place.:ugh:

Rod1

[Jan Olieslagers]

Sorry Rod, I really did not want to offend anyone. Still, even if the PFMS people started out doing the thing I am considering, they decided to go another way. That's up to them, they have all my respect and they certainly realised very nice stuff.
Perhaps I will follow their logic one day. Up till then, my efforts will just serve myself, apparently. Too bad.

Afterthought: even if PFMS started out willing to share their work freely, they choose to run in non-free O/S. So no, their project could never have been mine - again meaning no disrespect to their merit. Yes, you may call me religious. Happy flying.

[Gertrude the Wombat]

For myself I am stupefied to see almost everybody relying blindly on commercial stuff.
Personally I don't, other than Pooleys. And if I'm not convinced by what I read there I go and check in the AIP or the airport's own web site. As others have indicated I am hoping and expecting that this process would count as having done due diligence.

But I'm not convinced that using some freeware web site or software for weather or NOTAMs would count as due diligence, so I go to the Met Office and AIS web sites.

It's all a matter of risk assessment, and each pilot makes his own judgement (cf the common repeated threads on SEP IMC and night flying). Mine is as above. Others will come to different conclusions.

[IO540]

But I'm not convinced that using some freeware web site or software for weather or NOTAMs would count as due diligence, so I go to the Met Office and AIS web sites.

It's an interesting point.

It's hard to know where the UK CAA stands on this kind of stuff (third party notam sites which attempt lat/long parsing and a graphical presentation, and the countless weather websites which nearly all present data from GFS).

They are still mostly denying the existence of the internet - except that preflight briefing is de facto now impossible without it. They certainly avoid any mention of GFS; all the PPL training stuff is TMK purely UK MO material (F214, F215, MSLP).

There is now a totally bizzare gulf between the material (and the environment) presented to a pilot during training, and the material (and the environment) which the pilot will be working once flying for real.

There are no CAA prosecutions I know of in this area (what constitutes due diligence for preflight) and I am damn sure they would be extremely reluctant to try one, because given the PPL training syllabus a defence lawyer with more than one braincell would wipe the floor with them.

[Rod1]

“They are still mostly denying the existence of the internet - except that preflight briefing is de facto now impossible without it.”

No, that is not the case. It is possible to phone up and get a weather brief (at a cost). It is possible to phone up and get Notams. It is possible to work a plog out without the Internet. You can still fly without ownership or access to a computer. I would not wish to do it, but I know people who do.

Rod1

[IO540]

and you can eat soup with a fork, sure.

I am also sure the UK MO still run their faxback numbers... great stuff :)

[gpn01]

@Jan all credit for trying to apply some technology to overcome some problems (although I also accept there's a risk of trying to develop technology based 'solutions' to problems that either don't exist or that have been resolved). My techie proposal is really easy to design and implement. Write a non-OS specific app thast flashes up on the screen a really important message..... "LOOK OUT OF THE WINDOW!".

[soaringhigh650]

There is now a totally bizzare gulf between the material (and the environment) presented to a pilot during training, and the material (and the environment) which the pilot will be working once flying for real.


I have an idea - why don't you present your knowledge to them so they can improve their syllabus?

Or are some of these guys so risk adverse they can't or won't take change without another 10 years of paperwork?

[LH2]

Probably because they [OpenStreetMap] ripped it off high quality (by which I do not mean TomTom which uses databases several years old) commercial products.

Ah, but my esteemed colleague, you are very wrong there :)

It is a vastly different model to anything we would have known 10 or 15 years ago so it is not easy to get one's mind around it (even for me, and I have been involved in free software since 1995). However, there is a tightly controlled process which keeps copyright infringement to a minimum (I do not think you can say you are completely free of copyright infringement, as that's very muddy waters, but that's a problem commercial business face just as much).

Incidentally, there is a degree of cooperation between OSM and local and national mapping agencies (the Ordnance Survey for one, strangely enough).

Caveat: considering my background I'm probably much less familiar with OSM than I should be.