Skip to content

Commit 3281f8b

Browse files
authored
Merge pull request pebble-dev#63 from pebble-dev/posts/shiny-future
Add Shiny Future post
2 parents ddb2da9 + fe99d95 commit 3281f8b

File tree

1 file changed

+231
-0
lines changed

1 file changed

+231
-0
lines changed
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,231 @@
1+
---
2+
layout: post
3+
title: "The shiny Rebble future: one year after the Pebble server shutdown"
4+
date: 2019-07-29 18:00:00
5+
# categories: community
6+
---
7+
8+
Hi out there, Rebble world! [Joshua](https://joshuawise.com) on the
9+
microphone here for a bit. It's been a year since the Pebble servers have
10+
shut down, and I'm still wearing my beloved Pebble 2 on my wrist every day;
11+
as far as I can tell, it's *still* the case that nothing else out there
12+
compares to these things. As the buttons start to wear out on this one,
13+
though, I'm starting to ask some of the same questions that I hear echoed
14+
out in the Pebble world: what does the future of Rebble and of
15+
Pebble-like things look like? Well, I'm an optimist on that -- and I wanted
16+
to fill you all in on some of the reasons why. I'll catch you after the
17+
fold.
18+
19+
<!--more-->
20+
21+
Lingering in the back of my head have been three things: money, movement,
22+
and hardware. Without money, the Rebble web services can't run; without
23+
forward movement, we fade quietly into the sunset; and without hardware,
24+
there's nothing to put on our wrists! On all three of those, I have pretty
25+
darn good news in the short term, and I have ideas for the long run, too.
26+
I'm going to get into each of those in a moment, but before I do, I want to
27+
give you the same proviso [that Katharine did last
28+
year](http://rebble.io/2018/02/15/rebble-web-services.html): I can't guarantee the future, but I'm going
29+
to give you my best guess. Let's start!
30+
31+
## Money
32+
33+
I think this is the biggest question on everyone's mind after we watched the
34+
downfall of our beloved Pebble Technology Corporation: how can we build a
35+
_sustainable_ way to keep our watches ticking? After all, web hosting can
36+
be pretty expensive for a cloud application like Rebble, and we're not even
37+
making any revenue by selling watches. The solution that we came up with
38+
was to have a two-tier model: everybody gets access to the basic services
39+
for Pebble for free -- firmware updates, the app store, account management,
40+
and such, but the things that cost us money -- dictation and weather -- we'd
41+
charge a small amount of money for, and if we got the math right, the paid
42+
users would subsidize the free users.
43+
44+
So the question is: how well did that work out? I'm happy to say that it
45+
seems to be working pretty well. We have around 7,000 paid users, which
46+
comes out to about an annual run rate of around $190,000. This means that we're
47+
doing a fair bit better than breaking even for now: even though we're not
48+
drowning in cash, Rebble is definitely managing to pay its own rent from
49+
Amazon. This is heartwarming news, and I hope that it makes everyone feel
50+
better about the sustainability of Rebble itself, and also about Rebble's
51+
model for running services.
52+
53+
On that front, I wanted to thank all of you who are Rebble subscribers:
54+
without you, it wouldn't be possible to keep this service running, and you
55+
all are amazing proof that people will pay to sustain technology that they
56+
want to have more of in their life. :heart_eyes: [Maciej Ceglowski](http://idlewords.com/),
57+
proprietor of a small bookmarking service called [Pinboard](https://pinboard.in), wrote [a
58+
good piece a few years ago](https://blog.pinboard.in/2011/12/don_t_be_a_free_user/) about the importance of small paid services, and
59+
how they help build business models that center their users as a priority,
60+
instead of business models that center investors or surveillance systems as
61+
priorities; I have looked to that as inspiration, and I'm glad that we can follow in those footsteps. It's a joy to serve you all!
62+
63+
I suppose this would be the right time, also, to invite those of you who
64+
haven't joined us as subscribers to come on board! I think that $3/month (or
65+
$33/year, if you subscribe for a year at a time) is a great value for being
66+
able to dictate messages while I'm out on my bike, and for knowing at a
67+
glance when the sun is going to close me out without having to pull out my
68+
phone -- and, of course, as I noted above, subscriptions go towards keeping
69+
Rebble sustainable. So it's never too late to [join the
70+
Rebble-lution](https://auth.rebble.io/account/])!
71+
72+
## Movement
73+
74+
One worry, of course, is that if Rebble is standing still, we start to look
75+
pretty dead. After all, we'd been pretty quiet for about a year after the
76+
initial Rebble Web Services launch. Well, I assure you that we're not dead!
77+
Some of you with eagle eyes :eyes: have been watching commits to the Rebble
78+
repositories, and have noticed that the pace of development has picked up
79+
some over the past few months, culminating with the most recent [launch of
80+
Timeline](http://rebble.io/2019/07/24/its-timeline-time.html). You might be asking: what lead to this sudden surge in motion?
81+
And will the development continue?
82+
83+
Well, in May, [Katharine](https://twitter.com/KatharineBerry), [Ish Ot
84+
Jr.](https://twitter.com/IShJR), and [I](https://joshuawise.com) got
85+
together in a Hangout, and we spent some time thinking about the future of
86+
Rebble. We decided that we had a little bit of money left to spare, and
87+
that we wanted to reinvest it in the development of Rebble, and so we'd pay
88+
someone to do a little engineering work. Well, precisely, we'd pay, well --
89+
me! About a year ago, I left my previous day job to [do some software and
90+
hardware consulting](https://accelerated.tech), and as a result of that, it
91+
suddenly became possible for me to dedicate time in small quantities, rather
92+
than as "everything or nothing". For the past few months, I've been doing
93+
about six hours of paid work a week or so on Rebble (and maybe another five
94+
or six during the evenings and on the weekends!) to knock off some features
95+
that we've been hoping to build.
96+
97+
I've been doing a bunch of work behind the scenes on building development
98+
tools and administrative features so that we can help you out with billing
99+
problems, but Timeline was the first major fruit of that work that was
100+
visible to you, our beloved users. I don't think Timeline will be the end,
101+
either: on my to-do list is to start building out the app store developer
102+
portal, and even some work on longer-term ideas for keeping the Pebble
103+
ecosystem alive. Of course, all of the work that I'm doing is and always
104+
will be open-source, just like the rest of Rebble (it's even written into
105+
the contract!).
106+
107+
I'm doing this all on the back of the amazing work that Katharine did to get
108+
Rebble Web Services off the ground. She basically put this whole thing
109+
together single-handedly, and I owe her incredible thanks for such a strong
110+
foundation to keep building on top of. If you haven't watched her talk from
111+
[!!Con West 2019](http://bangbangcon.com/west), it's a great story about how
112+
we got here, told expertly in 10 minutes -- [go watch it; you won't regret
113+
it](https://www.youtube.com/watch?v=6qrfjbb5jKw)!
114+
115+
A year ago, Ish Ot Jr. wrote and said that we're just getting started, and
116+
he was right. It feels to me like we have a path forward to keep building
117+
Rebble into the future: we took a breather after we launched, but we're back
118+
at it, and now we're geared up to -- as Pebble liked to say -- keep making
119+
awesome happen! :rocket:
120+
121+
## Hardware
122+
123+
The last piece of the puzzle, I claim, is hardware. I'm not the only one
124+
with holes bursting in the buttons of his Pebble 2. Pebble Technology
125+
Corporation isn't making any more of the things, so in theory, our userbase
126+
is only shrinking from here, and that's not a good place to be. How can we
127+
keep more Pebbles running for longer, and is there ever a possibility of
128+
there being more?
129+
130+
### Pebble life support: you can save your Pebble now!
131+
132+
I am a die-hard believer in my Pebble 2, but as we all know, these things
133+
have a serious design defect: [the silicone buttons on the side fall apart
134+
after about a year](https://m.imgur.com/a/98Z7A). There have been a handful of makeshift solutions out
135+
there, but I never found the Sugru approach to be great. Fortunately, the
136+
quality of 3D printing has gotten quite good in recent years -- and even
137+
more fortunately, two wonderful Pebblers have been experimenting with
138+
hardware to rebuild Pebbles stronger, and, frankly, I think even
139+
better-looking. Tation and Astosia have [a Shapeways store](https://www.shapeways.com/shops/astosia?section=pebble&sort=) with all manner
140+
of exciting fixes for your Pebble 2, including entirely new cases that use
141+
the original "OG" Pebble buttons; they also have [Imgur albums](https://m.imgur.com/gallery/cvdagKI) [with the guts](https://m.imgur.com/a/nGeXEq0) of how to dissasemble and reassemble the devices. They're also experimenting with
142+
replacement 3D printed rubber buttons that you can glue right on. So if
143+
your Pebble 2 is falling apart, don't despair: there are fixes that you can
144+
make today to bring it back to life! There are also some [models available](https://www.thingiverse.com/thing:3464868) [on Thingiverse](https://www.thingiverse.com/thing:2599551), for those inspired to print their own.
145+
146+
The batteries in these things also don't last forever, we've found.
147+
Luckily, it seems like you can get anything in China if you know where to
148+
look -- people have had success with batteries from Aliexpress for the
149+
original Pebbles, for Pebble Time and Pebble Time Steel, and for Pebble 2.
150+
If you're handy doing a little soldering, Pebble batteries are not too
151+
tricky to replace, and you can get years more of reliable service from your
152+
beloved Pebble.
153+
154+
### Pebble life support going forward
155+
156+
Here's the bit where I start to speculate a little bit. I've heard
157+
[rumblings of the GadgetWraps guys making some side buttons for Pebble 2](https://www.reddit.com/r/pebble/comments/chqa9v/pebble_2_side_buttons_should_we_makesell_these/) with
158+
their beautiful silicone molding process. If they do, it could be possible
159+
to get the same lovely smooth feeling buttons that you were used to, without
160+
any major surgery at all. If that sounds like something you'd like, you
161+
might want to shoot them some mail telling them that you'd be interested!
162+
163+
I've also been talking a little bit with a low-volume injection molding
164+
company. I don't want to say too much about that, since I don't know where
165+
it will end up, but it's definitely possible that in the not-so-distant
166+
future, we could have new cases for Pebbles in all manner of interesting
167+
shapes designs, and materials. The logistics of selling them are very scary
168+
to me, but if there's interest, Astosia, Tation, and I are interested in
169+
making it happen!
170+
171+
### The post-Pebble problem
172+
173+
Of course, this is all well and good, but how can we bring new Rebblers into the
174+
fold? In some ideal universe, it would be nice to sell new hardware,
175+
without all of the pain of trying to limp along something that we didn't
176+
design. What if we could improve on Pebble with new features? After all,
177+
Bluetooth microcontrollers have come a long way in the last few years in
178+
terms of power consumption...
179+
180+
It sounds like a pipe dream, but it's not entirely. Over the Christmas
181+
break last year, I [designed an nRF52840-powered microcontroller board](https://github.com/jwise/aWatch2-hw) that,
182+
hopefully, could be shrunk down into the form factor of a watch. A month or
183+
two later, I had a circuit board that I called "Asterix" that I had
184+
Bluetooth and a display up and running on ... and a month after that, I
185+
managed to bring RebbleOS up on that, along with my Pebble app "Dali Clock".
186+
For the first time in history, I think, I had an application that was built
187+
using only Pebble tools running on non-Pebble Technology Corp hardware!
188+
189+
I don't want to get your hopes up too much. There's a lot of work to do.
190+
Barry -- who some of you on Discord may know as
191+
[ginge](https://github.com/ginge) -- did an *amazing* job laying the
192+
groundwork for RebbleOS, but as everybody who's come and asked about it
193+
knows, we still have a long way to go on the software front. The hardware
194+
that I have is about four times too big; I think it's definitely possible to
195+
shrink it, but it'll require a good amount of effort to do. And the
196+
question of manufacturing it looms huge in my mind, and I don't know where
197+
to even start with that.
198+
199+
But it feels possible, in a way that it didn't a year ago. I'm hoping that
200+
I can put some time towards that once I get Timeline settled in and under
201+
control. I don't know anything about manufacturing, and distributing and
202+
selling them scares the dickens out of me even if I manage to make more than
203+
one of them. So, hey, if you know about these things and want to take
204+
charge of a piece of it -- come in and help out! I'd be happy to work with
205+
you!
206+
207+
On that front, I'll leave you with an [exciting little video of Asterix
208+
booting into RebbleOS](https://www.youtube.com/watch?v=Gd2VBVRp234). It's
209+
way less complete than it looks: it's incredibly unstable, and it doesn't
210+
even know how to talk to the Pebble app yet. There's so much I don't know,
211+
and I don't even know how to estimate how much is behind it. But it's hard
212+
not to look at it and grin. What is the purpose of the future, if not to
213+
give us something to aspire to?
214+
215+
## Conclusions
216+
217+
Rebble is alive and well. I touched on three pieces: money, progress, and
218+
hardware. These three things all feed into each other: money keeps us
219+
alive, and helps us make progress; progress keeps us alive, and helps us
220+
make hardware; and keeping hardware alive keeps us alive, and lets
221+
Rebble keep making money. In the short term, Rebble is healthy on all three
222+
of these fronts. There's a plausible idea for how we could remain healthy
223+
going forward, too! It's going to take work, and it's going to take the
224+
continued support of you, our amazing community, but it's not impossible.
225+
226+
On a personal note, it feels very rewarding to get to serve yinz Rebblers.
227+
Pebble built a passionate group of users. I'm excited.
228+
229+
Ok, that seems like about it for today. Thanks for staying with us so far,
230+
and I'm looking forward to coming with you all into the Shiny Rebble Future
231+
-- keep your Pebbles on your wrists!

0 commit comments

Comments
 (0)