Did I Miss the 811 Software Fix?

[hemway]

Did I miss the 811 software fix that was supposed to be out at the end of the month? On the last Technical Chat, I thought they said that they were going to release 'at the end of the month' some 811 fixes, and some more a few weeks later.

Well it’s the END OF THE MONTH!!!!! Where’s the fix?????

Sorry, lost my mind. This is Dish we are talking about and they never elaborate on which month they are referring to.

 

[WLong]

Well, I havent been home in a few days so I am not sure if any updates have been applied to my box. As of friday, no there was not an update yet.

What version of the software do you have listed in your System Info?

Thanks,
Bill

 

[hemway]

I'm at P263 like everyone else.

My point was, once again Dish fails to deliver on what they said they were going to do.

 

[WLong]

The month isnt over yet. Still a few hours left.

 

[WeeJavaDude]

Personally I think Dish should stop saying anything about release dates.... Like I have said before, software delivery is an art and not a science. Believing in a firm release date is like believing there is an Easter Bunny. Differnence between D* and E*, D* does not talke about release dates so there is nothing to get dissappointed in when they don't happen.

IBM for example.. Does not talk about release dates and if they do they talk the the quarter granularity. They do this because they don't want to raise expectations. Microsoft on the otherhad constently talke about them and break them all the time. Just two ways of doing business. Question is do you want to know that something is coming down the pike and have them possible miss the date or not know anything and be pleasently suprised when a release appears?

I would suggest holding off a week or two before organizing a mob.

 

[hemway]

I guess my point is this, they used the 'at the end of the month' promise in December, in January, and again in February. That is a quarter of a year.

Add the fact that the 811 was supposed to originally be made available in August. They have had plenty of time to work on the software. Going back to an earlier post I put up here, they took the software from the 6000 and tried to get it to work on the 811 instead of writing the code fresh. Using your IBM and Microsoft comparison, it’s like trying to run Windows on a Mac.

Dish was looking at putting money in their pocket instead if intrusting confidence it its customer base.

 

[normang]

If your receiver was in standby and the code has not changed, you can't miss it. Its either available to download or its not.

Software development infortunately is a moving target, you either hit a date because things workout and you release it, or you test a release and find a problem more annoying than you originally had and you fix it, date slips. It will be here when it gets here and not before.

This stuff reminds me of my kids on vacation, "are we there yet?" or "How much longer" The answer is always the same, We'll be there when we get there...

Patience is a keyword when it comes to software. I would rather wait some extra days for a stable release, then to see all the whining and moaning (which will probably happen anyway) over a release that perhaps has more issues than the previous one because they didn't test it enough. I am sure that the next release does not address everything.. The last tech chat indicated at least two releases in the next several weeks for various known 811 issues.

 

[melvis1907]

I spoke with someone who claimed to be a senior tech support guy last night. (After lockup once a day for the past 10 days, I got fed up again) Anyway, I asked if he had any update on when the software fix was coming, and he said they did not know but that they were working on it. (Anyone heard that?) I blew-up on the guy and said they have been telling me the same thing since December and that I was getting tired of hearing excuses. He tried to change the tack and asked what problems I was having and I said - well look at your list of know issues, which must be multiple screens worth now and apply them all.
To make a long story short, they have no idea when all this crap is going to be resolved and sounds like the CSR's and teckkies are getting tired of hearing about it. I wish I didn't live in a Pegasus coverage area.

 

[AcuraCL]

... I would rather wait some extra days for a stable release, then to see all the whining and moaning (which will probably happen anyway) over a release that perhaps has more issues than the previous one because they didn't test it enough. ....

Hear hear.

 

[BudShark]

Patience is a keyword when it comes to software. I would rather wait some extra days for a stable release, then to see all the whining and moaning (which will probably happen anyway) over a release that perhaps has more issues than the previous one because they didn't test it enough. I am sure that the next release does not address everything.. The last tech chat indicated at least two releases in the next several weeks for various known 811 issues.

I agree with you to a point. At the point that the software does not perform where it should at release, they have failed. The problem is that it appears the system was released too early, meaning there are such core issues they have to rewrite substantial portions of the code.

If they were like any other quality run company, they would release the system with a very stable code, and then fix the small issues. There is only one reason why we have not received small incremental fixes - the software code base is bad. Otherwise we would have gotten a fix that addressed the OTA channel mapping, then another that fixed the 2day guide, and another that fixed the darkness issue, and another that fixed... etc. But instead we wait for a massive fix that is probably a near rewrite.

The other thing that KILLS me with Dish is the excuses. In the case of Dish it is NOT software as usual. They own the hardware development, the software development, and the transmission equipment. There is NO variability. It is inexcusable to have these issues and claim its 'software'. In the case of Dish it is ENGINEERING. It is a single platform with no variability. The TVs we plug into and the type of phone lines we have do NOT affect the software stability. ONLY their hardware, software, and transmission equipment affect it (of course, PSIP and 8VSB variability come into play - but thats not the extent of the problems we are seeing)

OK - stepping off my box until someone makes another excuse for bad engineering.

Chris

 

[normang]

I agree with you to a point. At the point that the software does not perform where it should at release, they have failed. The problem is that it appears the system was released too early, meaning there are such core issues they have to rewrite substantial portions of the code.

If they were like any other quality run company, they would release the system with a very stable code, and then fix the small issues. There is only one reason why we have not received small incremental fixes - the software code base is bad. Otherwise we would have gotten a fix that addressed the OTA channel mapping, then another that fixed the 2day guide, and another that fixed the darkness issue, and another that fixed... etc. But instead we wait for a massive fix that is probably a near rewrite.

The other thing that KILLS me with Dish is the excuses. In the case of Dish it is NOT software as usual. They own the hardware development, the software development, and the transmission equipment. There is NO variability. It is inexcusable to have these issues and claim its 'software'. In the case of Dish it is ENGINEERING. It is a single platform with no variability. The TVs we plug into and the type of phone lines we have do NOT affect the software stability. ONLY their hardware, software, and transmission equipment affect it (of course, PSIP and 8VSB variability come into play - but thats not the extent of the problems we are seeing)

OK - stepping off my box until someone makes another excuse for bad engineering.

Chris

Chris, I don't dis-agree, you have to wonder why these issues were not spotted right away, and you have to wonder why it takes weeks and weeks for updates. Some of it could be man-power related. Maybe They are using same engineers for all products and the 921 issues because of the cost of the unit are taking priority over 811 fixes.

While it would be nice if Dish was more forthcoming on things and explained clearly on their website what the issues are and what it takes to fix them, and why they take so long, it would be easier to understand. However they are not doing that perhaps for competative reasons, silence to us is annoying, but it doesn't advertise problems.

They do announce they are working on things and we wait impatiently for them to fix our favorite glitches, while others that own other products wait for theirs.

 

[bsic]

Not to get on a rant about software release management, but the way my company does it (who releases 4 rev's a year, and never misses schedule):

- Have a code-freeze date set, at least 4-6 weeks in advance of ship date. If you dont make the code freeze, your code is out until the next build in 3 months.
- Run regression/beta testing (i would think 2-4 weeks is enough for dish on this)
- Use a good revision control system to back out buggy code at the last minute (ClearCase?)

Seems pretty basic...we manage to get well over $100k per license of some of our titles. But dish's approach seems to be:

- Set a date
- Cram to get your code checked in by the day before, and hope it works.
- If the code has bugs, you delay your release and try to fix it. Set a new date.
- Cram to get your code checked in by the day before the new date, and hope it works.
- so on and so on

 

[WeeJavaDude]

Not to get on a rant about software release management, but the way my company does it (who releases 4 rev's a year, and never misses schedule):

- Have a code-freeze date set, at least 4-6 weeks in advance of ship date. If you dont make the code freeze, your code is out until the next build in 3 months.
- Run regression/beta testing (i would think 2-4 weeks is enough for dish on this)
- Use a good revision control system to back out buggy code at the last minute (ClearCase?)

Seems pretty basic...we manage to get well over $100k per license of some of our titles. But dish's approach seems to be:

- Set a date
- Cram to get your code checked in by the day before, and hope it works.
- If the code has bugs, you delay your release and try to fix it. Set a new date.
- Cram to get your code checked in by the day before the new date, and hope it works.
- so on and so on

Well I am not sure how Dish does release management so I will not comment on it. The way you describe your cycle is the way that release managment should occur, but ofcourse there are always those exceptions and when you do multi-platform development and share resources that can complicate things and the result is poorer released software.

Not all developement is the same and different models work better for different type of software. Also you have those last minute marketing needs that creep into the software at the last minute than can result in instability. I have seen a one line wonder that looks benine cause huge instability issues and removing bad code does not always bring back the code to its orginal state. There are hidden dependencies and since we are human some mistakes do get made.

I don't know what Dish's developement environment is like.... Based on what I have seen from the final product, I would say it needs a revamp. That must come from the top and it must be done in such a way as to keep Dish Profitable and meet customer needs. Not an easy task indeed.

 

[bsic]

but ofcourse there are always those exceptions and when you do multi-platform development and share resources that can complicate things and the result is poorer released software.

QUOTE]

... but I dont take that as an excuse since we have hundreds of engineers working on shared modules and nine OS's to support

They need a re-vamp in SW engineering..... as evidenced solely by the poor initial release of the 811 sw alone. As and end-user Im not really bothered by the bugs. As an engineer Im baffled as to how they could release with this quality! :shocked

 

[WeeJavaDude]

but ofcourse there are always those exceptions and when you do multi-platform development and share resources that can complicate things and the result is poorer released software.

QUOTE]

... but I dont take that as an excuse since we have hundreds of engineers working on shared modules and nine OS's to support

They need a re-vamp in SW engineering..... as evidenced solely by the poor initial release of the 811 sw alone. As and end-user Im not really bothered by the bugs. As an engineer Im baffled as to how they could release with this quality! :shocked

Guess I should have explained this piont a bit more. Basically I was saying that depending on resources and projects people may be pulled from different groups as needs arise and this can lead to poorer software quality. This is mainly a man power issue and trying to do more with less. What runs through my mind is this. The basic PM triangle.

Resources... Time..... Features. Reducing any one of these means that the others must increase. IF you want to pull a schedule in you most decrease features or increase resource. What I have seen is that this triangle is tossed out and End date management is employed. When this occurs, corners are cut and you have a higher defect count. Simple Engineering Management, but one that a lot of companies forget.

I was not trying to make excuses for Dish.. I agree with you that something is not right and needs correcting. Without any insite into the company, anything is a guess at best.

By the way.. I have worked in a software organization with thousands of engineers, multiple product release trains, and multi-site development. Managing at this level to produce quality code is no easy feat.

Cheers,

Ron

 

[John Quaglino]

Chris, I don't dis-agree, you have to wonder why these issues were not spotted right away, and you have to wonder why it takes weeks and weeks for updates. Some of it could be man-power related. Maybe They are using same engineers for all products and the 921 issues because of the cost of the unit are taking priority over 811 fixes.

The 921 has its own development team out of England.

 

[AcuraCL]

but ofcourse there are always those exceptions and when you do multi-platform development and share resources that can complicate things and the result is poorer released software.

QUOTE]

... but I dont take that as an excuse since we have hundreds of engineers working on shared modules and nine OS's to support

They need a re-vamp in SW engineering..... as evidenced solely by the poor initial release of the 811 sw alone. As and end-user Im not really bothered by the bugs. As an engineer Im baffled as to how they could release with this quality! :shocked

I dunno, perhaps they succumbed to the pressures exerted by ranting customers constantly bombarding them with threats to take their business elsewhere if they don't release this box immediately/fix *** (insert favorite 811 problem)/whatever. Hmmm. Interesting thought.

(We have met the enemy, and he is us.)

 

[Satch0922]

here is what they said to me today:

"Thanks for your email; Dish Network is aware of the issues with the 811
receiver. We are working hard to fix these issues and will have fixes
for
most with our next software release scheduled for today 3/2/04.

Thanks

C.J. Meany
Customer Resolution Specialist
Executive Office of Dish Network"

 

[hemway]

I powered off my 811 tonight and I was able to get the upgrade.

Fixes I noticed is the guide now goes out at least 24 hours, I stopped there, and it looks like the video in guide is working.

I did notice that the SD output is also w little bit darker then it was before, so I guess I'll just have to brighten up the SD port a little more.

 

[normang]

What version did it install?