Discussion about

August 22nd 2011 2:33 am

Palm post mortem: what could have been

For those of you who are still following all of the webOS news, I thought I'd share what I know about how things have been going at HP since the announcement, and shed some light on a few projects that never came to be.

Some of the information below comes from friends who work in the webOS Global Business Unit, née Palm. I've always been enthused about the potential of webOS, so with HP ditching all of its plans for webOS hardware, I was eager to pry some more details out of these people. Some of this info has also been floating around tech sites for a while, and I was able to confirm its accuracy.

When HP announced that they were ceasing production of webOS devices last Thursday, it all seemed rather curt and abrupt: their $1.2 billion Palm acquisition, dismissed with a couple of sentences in a press release? I hear it wasn't how they wanted to handle the situation — rumor has it they weren't even going to announce this for several more weeks, if not months — but their hand was forced when someone leaked the news to Bloomberg. Just about everyone in the company was caught completely off-guard, including higher-ups at the webOS Global Business Unit who were left in the dark: www.businessinsider.com­/todd­-bradley­-and­-jon­-rubin... Even HP sales reps were still hawking the TouchPad to businesses that same day, unaware of the surprise announcement.

Practically all software development resources at the webOS GBU were assigned to making sure that the TouchPad got out the door on time. Jon Rubinstein didn't think the product was ready to ship, but HP didn't want to wait any longer to release it.

webOS 3.0, the software version that ships on the TouchPad, was being developed literally until the day the tablet was launched. Engineers knew that the software was rough, and fragmentation was already starting to occur. webOS 3.0 isn't simply the same old webOS scaled up: HP essentially forked out a new version of the OS and engineered it specifically for the tablet form factor, with an entirely new software development framework (Enyo). This meant that the Pre 2 and Pre 3 were stuck on webOS 2.1 indefinitely, and wouldn't be seeing the benefits of 3.0 anytime soon — later confirmed by Rubinstein himself: https:­/­/developer.palm.com­/distribution­/viewtopic....

Speaking of the Pre 3, what exactly happened there? It was announced way back in February, but to this day it has yet to surface in the United States (and now it probably never will). Reportedly, the team responsible for the device hardware was working to fix several performance issues before the product could launch. Also of note, a new Touchstone Audio Dock which streamed music over Bluetooth was designed for the Pre 3, and was intended for release at the same time: www.engadget.com­/2011­/08­/17­/hp­-touchstone­-audio­-do...

Ever felt like Palm's phones looked slightly dated? Product development took a while because everything was managed in-house, design cycles sometimes lasted too long, and limited resources meant that development on one project would slow while employees were assigned to another. By the time new devices were ready for production, some of their designs were almost half a year old or more. That's a pretty long time in the rapidly changing world of smartphones. Carriers started to doubt their efforts, and by the time the Pre 3 was announced, Sprint had given up on them entirely. An HP employee over on Reddit provides some insight into the process: http://www.reddit.com/r/IAmA/comments/jo2oz/iama_hp_web_os_employee_amaa/

After the Pre 3, there were no more slider phones in the works; HP planned to supersede the Pre lineup with new slab phones. Several products were very close to market but got called off at the last minute — this seems to be an unfortunate trend for Palm. With all of the time spent working on cancelled products, they only dug themselves a deeper hole by setting back their design teams even further.

The first unreleased phone, codenamed "Windsor", was targeted for release late last year but was cancelled by higher-ups within the company. The device had a non-competitive design, the hardware team continuously missed their deadlines, and the project was reputed to be terribly managed.

The second unreleased phone, codenamed "Stingray", was slated for release at the beginning of this summer. It was a thin slab phone with a large glass high-resolution touchscreen, front-facing camera, rear 8MP camera, and no physical keyboard. Employees were excited about this phone as it would answer the cries for a more EVO-like webOS device. It made it all the way through the DVT stage — device verification test, one step before production — but too many software deadlines were missed towards the end, when engineers were pre-occupied with readying the TouchPad. By that point, the carriers decided they didn't want it anymore and the product was cancelled. Photo: www.precentral.net­/keyboard­-less­-webos­-phone­-sting...

A smaller 7-inch tablet codenamed "Opal" was already in the hands of the QA team for debugging, slated for release later this year (16/32GB, Wi-Fi/3G) — it was spotted in an FCC filing just weeks ago: www.geek.com­/articles­/gadgets­/hp­-touchpad­-go­-7­-tab... And last but not least, the Foleo was ready for a comeback: HP had webOS running on a netbook form factor internally by late Spring, with a new build of the software codenamed "Dartfish".

I don't feel bad for the HP executives, who lacked the ambition to properly execute; but I do feel bad for the Palm employees who worked incredibly hard in hopes of seeing webOS succeed. Hardware engineers are currently stuck looking for new jobs, with almost no chance of relocation within the company. While HP has stated that they will continue to develop webOS, members of the software team I've chatted with are basically considering it dead. I feel especially bad for the developer community, and the developer relations team who is doing their best to appease everyone's concerns while not knowing much more than anyone else. It seems like everyone is stuck waiting for more instructions from the top.

In an interview at D9, HP CEO Leo Apotheker said that he kept the Palm team shielded in their old Sunnyvale headquarters to protect them from the HP bureaucracy. Maybe letting the same managers and organizational structure continue to exist at Palm wasn't such a good idea, when it clearly didn't turn out so well for them in the past. Perhaps if HP had been more aggressive at taking control of these projects, there wouldn't have been so many setbacks. Then again, under Apotheker's reign, maybe HP wasn't willing to put much muscle into it at all.

Amidst all of this, the $99 TouchPad fire sale is proving immensely popular, with both online and brick-and-mortar retailers selling out of the device in mere minutes. webOS advocates within the company are sending this up the chain as their last-ditch ammunition to show that there's still lots of interest in the platform and great potential for a webOS community. Will this sway the minds of HP execs in any way, and reverse the platform's ill fate? Like webOS developers and Palm employees, we'll just have to keep waiting and see what happens.

sort by

8 replies

Wow, thanks for posting this. It sounds as though we missed out on some really interesting hardware. I am still hopeful webOS will stick around somehow.
2 like dislike

The fact that the software team internally feels it's dead is disheartening. I really do hope they continue to push the software since it'd be a shame to see it die. But given how they've executed so far, I'm not sure they can recover.

I just hope they don't somehow shut down the webOS internals group. They are the only ones left keeping us firesale buyers going for the next year or so.
2 like dislike

Whilst it's nice to have it in this more succinct format, I think it's incredibly disingenuous to effectively source 80% of this article from a single IAMA Reddit thread of a HP employee without credit. In certain parts, the sentences are barely even reworded: http://www.reddit.com/r/IAmA/comments/jo2oz/iama_hp_web_os_employee_amaa/

This article: "already in the hands of the QA team for debugging"
Reddit thread: "They are floating around with the QA teams that were debugging them"

This article: "...was cancelled by higher-ups within the company. The device had a non-competitive design, the hardware team continuously missed their deadlines, and the project was reputed to be terribly managed."
Reddit thread: "the first canceled product was done internally by the execs. It was a smart decision as the project was poorly managed from the start. The design was poor, it didn't have much direction, and the hardware team was constantly missing deadlines."
2 like dislike

As I stated, some details here have already been posted around the web, but I was able to confirm their accuracy with my contacts at the company. To your point, I've updated the article to include links where applicable.
5 like dislike

"webOS 3.0, the software version that ships on the TouchPad, was being developed literally until the day the tablet was launched."

Not true at all. The software that was shipped on the touch pad was frozen a month and a half before the release date. What else in the article is inaccurate?
0 like dislike

This is absolutely true. HP engineers were still working on a number of components targeted for the initial release that slipped, including deeper hardware acceleration. Because of the new 3.0 development framework, engineers were also working on system applications that were rewritten and had not yet been completed.
1 like dislike

That is a good point, it would have been impossible to distribute otherwise. I don't recall any updates on the TouchPad launch.
0 like dislike

Awesome post, Aaron! Great read!
0 like dislike

20 users following this discussion, including:

  • phi
  • JonathanHoover
  • peterto
  • linh
  • Dpmt
  • cass
  • Janricr
  • dave
  • Willmonwah
  • aaron

This discussion has been viewed 10145 times.
Last activity .