Tesla Priorities: Refine Autopilot or Fix Everything Else?

[RogerHScott]

Is it too much to ask that a "finished product" should function as intended, whether it is tangible like a refrigerator or intangible like a computer program?

As a professional software engineer for many years, I could write you a book in answer to this (and your other) questions, but...
First problem: the "intended function" of virtually all software systems is woefully underspecified (for any number of reasons, including
the entity asking for it being too lazy to spell out exactly what they really want). Second problem: if you had a specification for a software
system comparable to the specification for a refrigerator, faithfully implementing it would cost far more than anyone would be willing
to pay. This is the basic problem with all of the whining about software quality; no one would pay for the level of quality they think they
want. Germane to this larger discussion, the market has shown again and again that people don't actually want higher quality, because
they choose new-and-shiny over not-so-new-but-polished time and time again.

Just how much work does it take to write these commands?

A lot. Not the numbers you suggest, but big enough numbers as to be pretty overwhelming.

Are there diagnostic programs available to check for errors in the original program?

This is actually my specific field. The answer is yes, kinda. There are both "static analysis" tools (my thing) that can find problems
in the programs themselves and "dynamic analysis" tools that detect problems in what a program does when you run it. Both
kinds of tools are very useful and any serious software dev. organization should be using them. But even putting the best of all
such tools together still leaves many problems undetected, particularly problems of the form "you meant it to do A but you programmed
it to do B". Or, worse yet, "your customers expect it to do B but you sincerely believed it should do A and programmed it correctly to
do just that".

Although I'm sure there are "bugs" in Tesla's code, most of what I've seen discussed here are really "feature requests" (even if they're
"how could you seriously leave this feature out?!" kinds of things). To me, at least, there's a huge difference between claiming to do
something but having done it poorly and not claiming to do something at all. The former is the sort of thing that eroded confidence in
American automakers in the '70s and '80s. Is anyone suggesting Tesla has a major problem in this area? The latter might eventually
get you left behind by competitors, but customers won't despise you.

 

[brooklynrab]

Roger, You make some good points, but the NAV system isn't a feature request -- it is a core system that works like crap. A speedometer that is not readable with sunglasses on is a safety threat.

 

[Boatguy]

Although I'm sure there are "bugs" in Tesla's code, most of what I've seen discussed here are really "feature requests" (even if they're
"how could you seriously leave this feature out?!" kinds of things). To me, at least, there's a huge difference between claiming to do
something but having done it poorly and not claiming to do something at all. The former is the sort of thing that eroded confidence in
American automakers in the '70s and '80s. Is anyone suggesting Tesla has a major problem in this area? The latter might eventually
get you left behind by competitors, but customers won't despise you.

I'm also a former software professional and I think you make an important distinction. I completely agree that much of what is there works, and a lot of the complaint is over the fact that working as intended is not the same as working the way the customer wants (though the occasional need to reboot evidences at least a few bonafide bugs).

Then there are a lot of feature requests. We can quibble about how high to set the bar, but I think most of the complaints here are about "features" that are pretty basic. Things that many customers feel should be there in a $100,000 car and that are there in other cars costing half as much (like matching driver profiles to key fobs).

Infotainment is poor. Period, full stop. Tesla either needs to replicate the functionality of CarPlay and Android Auto, or adopt those products. The later would be far better for everyone, including Tesla since it would deflect our future complaints to Apple and Google and Tesla can devote their resources to software that creates unique intellectual property, not reinventing the wheel.

Navigation is also poor. Driving an MS/MX over an extended distance frequently requires some planning, for which Tesla provides no support. It should be possible to plan a trip and evaluate alternative routes and schedules. Where do I want to spend how much time charging? What if the weather turns nasty? What if we decide we like that place and want to stop for the night? What if I want to drive 85mph? Ideally this can be done at home with something like evtripplanner and then downloaded to the car, but certainly it should be possible while sitting in the car. Route planning is more than just "get me from the car's current location to another location".

The list goes on and on and BertL did a nice job identifying and classifying the issues. Some Tesla product manager should take his list and start getting it done. Yes, it's work, but it's not work that requires Tesla's best and brightest, it's straight ahead s/w development, not much invention.

 

[AmpedRealtor]

The fact that Tesla has not added waypoints, multiple destination routing, multiple route selection, and other basic features to navigation tells me that it may be impossible or very difficult to implement. Tesla's navigation is a split system, relying on both Google and Navigon. Perhaps there's a problem making those two work together in the additional ways. I don't know, but it seems like a strange omission.

 

[BertL]

While I'm sure I'll get hate comments over this from some, and I'll probably regret making the post because I really don't ever like making anything personal...

My opinion is the whole issue with Tesla's lack of attention correcting some of the long outstanding firmware bugs and improving some of the Infotainment functionality many owners desire, is because Elon himself does not see them as important in his scheme of things. Let's not forget the man himself is a coder and started what PayPal has become, and if people have read Vance's book about him, and any number of other articles, I'm sure they walked away like I did that only he is in charge of Tesla and everything is done only his way. Given other wonderous things Tesla has accomplished, I just don't believe anything with Infotainment is really insurmountable even with the nuances required maintaining a real-time mapping display with offline Nav database. I agree that most of these non-Autopilot things we've discussed are not rocket science as @Boatguy and others suggest.

As discussed previously, I get the need for prioritization, especially in its early days when Tesla is not generating a profit ...but more importantly, I doubt very much Elon personally uses Nav, USB support, or say multiple drivers of the same personal MS every day himself (think FOB-to-Profile need), so even with how smart he is, he likely does not see those issues the same way a growing number of "real" owners do in our every day use. I'm not aware there is an evolving representative cross section of owners that provide an ongoing true advisory role to Tesla. Sure, some owners send in emails to Tesla with comments and suggestions, but really, are they used, and do those comments represent a true cross section like other more mature mfgrs have? Also, if a firmware change will not generate positive press, more immediately resolve an otherwise costly defect that effects Tesla's bottom line, addresses growing negative press, or is just flat out interesting to Elon to pursue as what he sees as his vision for tomorrow, I suspect it is not prioritized. I've seen little to suggest otherwise. We're still at the stage where there are more owners and shareholders (and of course 300K M3 reservations) that accept Elon's direction as the only right one, and who defend his every action -- lashing out at folks like me asking for just a little more consideration to make my $100K investment a bit more reliable and usable as my former $60K Lexus, BMW and MBZ were for years. I would appreciate acknowledged firmware bugs being addressed with the same diligence SCs do for hardware problems -- especially since MS is touted as being a "software-enabled" vehicle unlike any other. Additionally, Tesla has not (yet) gotten to the point with volume or competition where they have had to start utilizing firms like JD Powers to rate customer satisfaction that is more easily compared as apples-to-apples with their competitors, and likely impacts future sales to varying degrees (but the time will come.).

As such, IMHO there is little pressure on Tesla and Elon to open their aperture in terms of improving customer satisfaction that fixing a bunch of firmware defects or making a lot of functions more usable would provide to some owners -- even if that were done slowly but surely as I've proposed several times in threads here in TMC.

I bought into Elon's Tesla vision of what they are trying to accomplish in the long term by purchasing my own MS this past Fall. I sold my Lexus hybrid and MBZ ICE, going all-in to Tesla's vision to own my single MS. It's an amazing vehicle, but I will continue saying improved demonstrated balance is required by Tesla and Elon -- both enabling the future WHILE taking better care of today's owners like me that have knowingly funded Tesla's future through our investment in purchasing a MS or MX. Percentage-wise, M3 owners will not be making that large of future financial contribution with their $35K base entry price, so fixing some of this now for us that are early owners, also allows future Tesla owners including M3 to reap many of the benefits. Being so focused on HIS vision, I just don't believe Elon really gets the difference between where Tesla is today and what it will be like a few years from now with masses of owners and potential buyers that have more competitive alternatives in the marketplace, grousing about (a bunch of little) things Tesla could have proactively addressed ahead of time, instead of the more costly approach trying to catch up when the firestorm or negative reviews on "missing basics" ensue. It will happen. I only hope I personally don't have to wait that long, as I'd love to not have some of these things eat at me each and every day -- making me question what my next vehicle will be -- and get on with enjoying my otherwise technologically superior MS. I'd like to move back off the fence, being a full-bore advocate for the brand and MS, but it requires Tesla to step up, resolve firmware bugs no different than hardware failures, and start showing progress pulling Infotainment up to and hopefully beyond what the competition provides.

 

[AmpedRealtor]

As such, IMHO there is little pressure on Tesla and Elon to open their aperture in terms of improving customer satisfaction that fixing a bunch of firmware defects or making a lot of functions more usable would provide to some owners -- even if that were done slowly but surely as I've proposed several times in threads here in TMC.

Hey Bert, hope all is well! I have a feeling the above may change with Model 3. While many are Model S owners, many are not and they will want software that is at least comparable to what they currently have or on the cars that they are cross-shopping. Mainstream consumers tend to be less forgiving than most Model S owners who are willing to give Tesla a pass on a lot of things.

 

[BertL]

Hey Bert, hope all is well! I have a feeling the above may change with Model 3. While many are Model S owners, many are not and they will want software that is at least comparable to what they currently have or on the cars that they are cross-shopping. Mainstream consumers tend to be less forgiving than most Model S owners who are willing to give Tesla a pass on a lot of things.

All good here. We sent you some rain yesterday. Gorgeous morning now, with more rain tonight on its way.

...and yes, I agree. M3 will change the dynamic. I'm loosing confidence something will change nearer-term, and I know am sounding like a broken record, but selfishly hope Elon and Tesla get ahead of that curve for present owner's benefit, as well for as the longer term benefit to the brand. I want Tesla to be successful. Having spent all my career in software development, HW & SW tech support, & customer service -- including starting up remote technical support centers supporting both warranty and as a profit center -- and seeing how growth changes the dynamics, this whole subject just gets me going as you can tell. I learned the very important distinction between satisfied vs very satisfied customers (think 4-star vs 5-star), and what happens if you allow your customers to start feeling somewhat un-loved or their concerns not being attended to as they expect. If you're not careful, none of that is apparent or necessarily on your radar when you're in start-up mode and have an essentially unique product with customers willing to spend what it takes to have it, but is imperative for long term profitable growth as the dynamics evolve. I want more than anything for Tesla to understand that now and do something about it before it bites them in the you know what. I'm convinced we would all loose in that situation -- shareholders and owners alike.

 

[yak-55]

It boils down to ... "when does this (passionate) hobby mature into a (profit seeking) business?" Long term success requires the latter, although many early adopters will criticize it as "selling out".

Few businesses are successful in the long term ignoring their customers. In fact, I can think of none who succeed at this scale with the idea that "only he (the CEO) is in charge .... and everything is done only his way."

 

[msnow]

I'm finding it very hard to recommend Tesla to my friends anymore because of this. These aren't just little things. Many, like navigation, are essential to the driving experience but the larger problem of not fixing long standing issues shows a lack of organization, focus and the ability to prioritize none of which bodes well for the future.

 

[neroden]

To me, at least, there's a huge difference between claiming to do
something but having done it poorly and not claiming to do something at all. The former is the sort of thing that eroded confidence in
American automakers in the '70s and '80s. Is anyone suggesting Tesla has a major problem in this area?

Yes. Tesla has MAJOR problems in this area.
-- The Nav "turn by turn" is nonfunctional, suggesting stupid routes
-- The media player stutters between tracks, making it impossible to play an album like Sgt. Pepper correctly
-- Both problems have been present since 2012 with not one finger lifted to fix them. Both are worse than programs I can get for free on my computer.

In hardware, Tesla has done very well: they have delivered what they promised. In software? Not so much.

 

[ohmman]

Out of curiosity (and being completely serious/not snarky), which manufacturers have better navigation systems? I only have experience with the Honda/Acura system and it's horrible. Tesla beats it outright, no competition. I'm assuming most of you are comparing against better benchmarks.

 

[msnow]

Out of curiosity (and being completely serious/not snarky), which manufacturers have better navigation systems? I only have experience with the Honda/Acura system and it's horrible. Tesla beats it outright, no competition. I'm assuming most of you are comparing against better benchmarks.

The main problem for me is inaccurate turn by turn which is totally unacceptable. For the rest (waypoints, route choices, media) I'm comparing to Lexus RX, Mercedes.

 

[Boatguy]

Both my 2011 Mercedes and 2014 BMW i3 offer alternate routes.

 

[Boatguy]

It boils down to ... "when does this (passionate) hobby mature into a (profit seeking) business?" Long term success requires the latter, although many early adopters will criticize it as "selling out".

Geoff Moore wrote the book on this 20yrs ago, "Crossing the Chasm". Tesla has not yet done that. As many owners point out, what the early adopters are willing to ignore, the mainstream is not. Tesla will exhaust the M3 early adopters and encounter some far more critical reviews long before it fulfills those 300,000 refundable deposits.

I have an i3 for a "local" EV and it's far superior to the MS for local driving (better maneuverability, cargo access, etc), but it's not a touring car. I bought the MS for a touring car, or at least trips > 50 mi, which means I need excellent nav and excellent infotainment to plan and enjoy the longer trips.

 

[BertL]

Out of curiosity (and being completely serious/not snarky), which manufacturers have better navigation systems? I only have experience with the Honda/Acura system and it's horrible. Tesla beats it outright, no competition. I'm assuming most of you are comparing against better benchmarks.

I have owned Lexus (my DD for years), but also BMW and MBZ in the past 10 years - all with their upgraded level of available Infotainment system. Honestly, each Nav/Infotainment has quirks compared to the others, and each has people that like, dislike, or claim it's a piece of junk, but IMHO each had far more capabilities than what my MS does, and had less quirky bugs needing workarounds. My last Lexus RXh was a 3rd-gen 2013 and its visual interface was somewhat blocky like an older PC, but functionality was there across the board. If I had purchased a 4th-gen 2016 RXh instead of my MS, the interface was completely updated, much larger and visually has a HD appearance more like what we know on the MS. Of what I've owned, my 2014 MBZ was certainly the most refined of the bunch when it came to Infotainment -- visually and function-wise -- and had no bugs that I was aware of the day I sold it.

To @msnow's point, I've had out-of-the-ordinary Nav routing issues where I've been sent on a wild goose chase into the south 40 on each one of my former Nav, but I personally have not seen that (yet) on my MS. While different in each vehicle, each odd route was something I could recreate, so it was some quirk between the programming and Nav database I could workaround using waypoints or some other means. In my former Lexus, rather than get sent to a part of L.A. I had no business in and wanted to avoid ever accidentally being taken through again, I just set it as an area to avoid (oops, we don't have that on MS) and never was routed that way when I had to go up to the big city.

 

[RogerHScott]

Yes. Tesla has MAJOR problems in this area.
-- The Nav "turn by turn" is nonfunctional, suggesting stupid routes
-- The media player stutters between tracks, making it impossible to play an album like Sgt. Pepper correctly
-- Both problems have been present since 2012 with not one finger lifted to fix them. Both are worse than programs I can get for free on my computer.

I guess I just don't see these issues, annoying as they may be, as being comparable to the IMO more fundamental problems
that American car makers had in their bad years. But maybe I'm clinging to some sort of dinosaur notion that cars are, at the end of
the day, primarily about driving.

 

[msnow]

I guess I just don't see these issues, annoying as they may be, as being comparable to the IMO more fundamental problems
that American car makers had in their bad years. But maybe I'm clinging to some sort of dinosaur notion that cars are, at the end of
the day, primarily about driving.

True but driving is much more enjoyable if you can trust your nav, enjoy your music as you used to, have software such as Trip Planner come out of beta and work as expected, etc, etc, etc...

Cars are different things to different people. A true dinosaur may think they are just a form of transportation but you can get that for a lot less than $100k+. I'd just like them to fix the stuff they have out there now. I don't think that's asking too much.

 

[ohmman]

Of what I've owned, my 2014 MBZ was certainly the most refined of the bunch when it came to Infotainment -- visually and function-wise -- and had no bugs that I was aware of the day I sold it.

I had a MB C300 loaner from my Tesla SC once. I toyed with the infotainment system and found it cumbersome. It had a very odd interface controller mounted in the center console. Looking online, it appears that's called COMAND. I found that I had to drill down a lot, but that might have been a function of my inexperience with the system. Overall, I was pleased to get back to a touch interface and a large screen. MB calls their screen large, but it's not.

I agree that navigation has room for improvement. Turn by turn can do some weird stuff, especially if you let it reroute you due to traffic. Partly I imagine Navigon is to blame for that, and Tesla's to blame for using Navigon. I don't know for sure.

Overall, it works for me, but I can tell there are some people for whom it doesn't work. It's just surprising to me that someone would prefer that COMAND and the MB system, which to me feels very Windows 2000 with a trackball.

 

[BertL]

I had a MB C300 loaner from my Tesla SC once. I toyed with the infotainment system and found it cumbersome. It had a very odd interface controller mounted in the center console. Looking online, it appears that's called COMAND. I found that I had to drill down a lot, but that might have been a function of my inexperience with the system. Overall, I was pleased to get back to a touch interface and a large screen. MB calls their screen large, but it's not.

I agree that navigation has room for improvement. Turn by turn can do some weird stuff, especially if you let it reroute you due to traffic. Partly I imagine Navigon is to blame for that, and Tesla's to blame for using Navigon. I don't know for sure.

Overall, it works for me, but I can tell there are some people for whom it doesn't work. It's just surprising to me that someone would prefer that COMAND and the MB system, which to me feels very Windows 2000 with a trackball.

Agree with you that how one accesses functions, i.e. the UI, and number of touches and/or how you drill down through menu systems is a whole different discussion. COMAND (MBZ) took me a while to get accustomed to, and well, iDrive (BMW) took me even longer; Lexus was easiest for me as I owned 5 different models across 20 years, so their changes were incremental and evolutionary, not revolutionary even between models much of the time, like it was for me owning two different brands for so many years and trying to keep straight how to do things in each one when I switched driving different cars. I generally knew the function was there, but sometimes finding it was the challenge. Even with my MS, now that I've owned it for 6 months as my one and only vehicle, I still at times struggle with if what I want to check or change is under "Controls" or "Settings"... or "no, that's one of the Tesla Apps (like Nav), so I need to go there to change the setting", and I find myself occasionally touching all over the place on the 17" taking my eyes off the road for too long if I'm not careful. See what I mean, even with touch-screen UIs?

I think anyone jumping to a new brand will have the same learning curve when it comes to UI access, but if the basic functionality isn't there or is broken as I believe most of this discussion is about, I at least don't even consider how its best accessed and utilized.

 

[neroden]

Out of curiosity (and being completely serious/not snarky), which manufacturers have better navigation systems? I only have experience with the Honda/Acura system and it's horrible. Tesla beats it outright, no competition. I'm assuming most of you are comparing against better benchmarks.

Bluntly, I don't know about other manufacturers, but if I print out a set of turn-by-turn instructions from Google Maps at home before I leave, it's better than Tesla's turn-by-turn instructions. So, a paper printout is better.

With gapless media playback *and* this, what frustrates me is that it really shouldn't be difficult to do a decent job.
-- Hire a programmer to fix the media player. Look at the code for one of the open source players with gapless playback (like "audacious") as a reference; you'll be done in a month.
-- Tesla's already using Google Maps on the big screen, they can use Google turn by turn directions too.
There. Problems would be solved. They just haven't *bothered*.