In the "under the hood" discussion, the authors of Wave said that they
have an extremely fast method of joining individual edits, so they
probably use that for writing to disk, caching edits in memory and
using a thread to spin the input into a stream and write streams to
disk.
On Dec 5, 8:03 am,
Google has a conversation manifest document in the works for the
protocol as well, which will hopefully define a solid XML structure
for Waves, instead of the somewhat unknown one we have now (Robots and
people have no access to them that is).
On Dec 4, 7:13 am, "\\x00" wrote:
> You have the wave
Robots do not fire events for themselves. If a Robot submits a blip,
it will not receive a BLIP_SUBMITTED event about it. If a Robot
creates a new wave, it should not receive a WAVELET_SELF_ADDED or
WAVELET_BLIP_CREATED event either.
On Dec 4, 3:07 am, hvt_kg wrote:
> Hey guys,
>
> I had test and
Hehe, that would be a massive breach of security. There is no way for
a robot or a gadget to add itself to every public wave (or any, they
must be added by a user). There is no way to "who-is" a user to find
out what waves they are following, unless the robot is also a
participant in said waves. To
It seems that Wave currently has no way of "deleting" waves. If you go
to your "Trash" folder, you'll see all the waves you've tried to
delete via "trash." There may be a way in the future, but it will
probably be something like Gmail's trash with a few more rules.
On Dec 3, 9:42 am, fabrizio-mc
The scroll bar is quite effective in my opinion, it smooths out the
scrolling, shows where the page is currently, and looks pretty. It's
better than a normal scroll bar because on a large wave, that normal
scroll bar would be about 4 pixels high and impossible to click,
instead you always have a ni
Currently, all waves are stored on the server that hosts them. Google
is one of the more secure storage centers, but even they have flaws,
especially when it comes to confidential data. When Wave grows up
enough to be federated to any server, you can set up your own and run
the wave from there (jus
I've also not had WAVELET_BLIP_CREATED work for me, but if one keeps
an ArrayList of BlipIds and then checks on BLIP_SUBMITTED if the blip
id is contained in the arraylist, then one can accurately count new
blips by adding them to the list and keeping a counter that resets
however often after firin
oo, but there is a preliminary
one for consideration that hopefully gets the idea across.
On Dec 1, 11:21 pm, Olreich wrote:
> It doesn't seem like that gadget should be causing an onblipsubmitted
> every second...
>
> On Dec 1, 9:41 pm, Stephen George wrote:
>
>
>
> >
It doesn't seem like that gadget should be causing an onblipsubmitted
every second...
On Dec 1, 9:41 pm, Stephen George wrote:
> Daniel, I agree that it's difficult to remove a robot from a wave of
> over 200 participants.
>
> I like to use bouncy-w...@appspot.com in that situation to avoid
> hav
context.GetBlip().GetDocument().GetText() is the guess off the top of
my head. But basically, grab the blip from the event data
(context.GetBlip). Then get the XML document for that blip
(.GetDocument). Finally, grab the textual information of the Blip
(excludes all gadgets, pictures, and anything
As I understand it, the profile is made to store the profile
information of the robot, and capabilities.xml is made to tell the
wave server what events to send to the robot. But, I guess there was a
method of storing profile info in the capabilities.xml, but that's now
deprecated, so woohoo semanti
I thought the entire client/server relationship was JSON or XMPP...
On Nov 25, 6:43 pm, Trejkaz wrote:
> On Thu, Nov 26, 2009 at 12:32 AM, Venkat Polisetti
>
> wrote:
> > Currently Google does not support hosting Robots on servers other than
> > the App Engine. In the future it will, I hope.
>
>
There is a chrome extension to notify of new waves, I'd assume that
firefox also has a similar extension, and as Wave gets more beefy and
worked out, a wave notifier will probably come.
On Nov 25, 3:04 pm, Venkat Polisetti
wrote:
> Hi all,
>
> I have been building a Support Robot for wave for our
robot authentication information.
On Nov 26, 1:52 pm, Vikram Dhillon wrote:
> Just another question: how would the concept of a desktop based wave work?
>
> Regards,
> Vikram
>
>
>
> On Wed, Nov 25, 2009 at 12:50 PM, Olreich wrote:
> > Currently, robots are response
1) There is no client side API that I know of, though someone has
managed to create a google wave notifcation extension, so there must
be some way to do it.
2) There is no issue opened that I can tell, though you can certainly
start one, as a client side API would be quite useful for integration
w
Sounds dangerous, I would suggest being extremely careful about who
you invite to your waves all the time though, not just when something
could have issues. Waves are easily hijacked if someone adds a Robot
that will hijack it. But until an ownership model is implemented,
waves are not safe.
On No
Currently, robots are response oriented. They much respond to another
event to do anything. A Cron job may be able to run a createWavelet or
NewWave function without being added to a wave, but it is still an
event which the robot responds to. In the future using cron or another
timer mechanism to c
Robots can now see the entire conversation upon addition, by adding a
context="children" to the "WAVELET_SELF_ADDED event in
capabilities.xml according to cmdskp. I've not tested it, but from the
feedback it appears to work intermittently this is probably because
the whole child-parent thing and co
http://wave-robot-java-client.googlecode.com/svn/trunk/doc/com/google/wave/api/oauth/package-summary.html
And because I've never used that before, I'll allow others to tell you
how it works and if it's indeed the right package for authentication
on a wave for robots. But that will probably be a go
Kudos:
Useful for setting a range of dates and checking availability.
User friendly and simple to create and manage.
The idea of yes/no/maybe for multiple related instances makes great
sense.
Suggestions:
Change the font color or background color for the "no" state. It is
difficult to read
Looking at the structure of the Python API, it appears that context is
a variable that is specific to a certain request, to act upon that
request, you would need to pass the context variable off to the other
instances of the Robot in some fashion. If you can achieve that, then
just grab the right c
Related to the idea of agents vs. robots: What is a probable time-
frame for agents to be exposed on Google Wave? Is it Weeks, Months,
Years, possibly Never, or maybe Unknown?
On Nov 23, 11:33 pm, "pamela (Google Employee)"
wrote:
> Just wanted to make a few clarifications - Spelly is an agent, n
now
> since the API development is moving forward.
>
> On Nov 23, 3:22 am, Vikram Dhillon wrote:
>
>
>
> > The visual really helped, I wasn't thinking properly so I apologize for that
> > and thanks for your clarification on the subject. So from what I understand
&g
GetChildren on the root blip will get the first blip. Unfortunately,
because the root blip has to be the blip submitted to have that
function work, it's limited. In fact, it's even more limited because
of the fact that the root blips child's child is null in the current
version of the API. (an anno
Deleting the child of the rootBlip will not delete it's
"children" (the rest of the conversation). This may be a bug. Of
course, I'm guessing that his application wants that update to be
immediate, so that he can post other data in the blip and then allow
people to respond to it. He could lock the
The parent/child system in the API is currently shot. I hear that
there's a new API going into place sometime to get all the blips in a
wave. I don't know the implementation date, but it's in there.
On Nov 21, 4:20 am, neolinux wrote:
> No solution?
>
> On 14 Nov, 00:18, neolinux wrote:
>
>
>
>
Maybe it's a bug, maybe they did it on purpose. I'm not sure, as I can
see both an "automatic" default value and "null" default value to
work. but, to make this fix simple, just throw a default value into
the constructor statement of your RADIO_BUTTON_GROUP.
To file a bug: http://code.google.com/p
If you want to do it with Robots, you may get the functionality soon,
as it looks like the Wave team is building a version of the API where
the Robot can send messages, at least to get context of blips, this
may mean good things for the Robot sending messages and having text
appear incrementally (y
in the future in my first reply -
> nothing about purpose or that you're not supposed to, etc.
>
> On Nov 20, 4:10 am, cmdskp wrote:
>
>
>
> > I really should read the rest of a thread before jumping to reply to
> > someone asking for an answer in a link above.
With that code, you are most likely getting a NullPointerException, as
blip has no parent as far as your program is concerned. This is an
issue with the context of events, as whenever the blip in question is
not the e.getBlip() there is no parent and no children. It gets quite
annoying.
However, C
Look at the Docks, they'll tell you the different functions associated
with each of those classes, but pretty much anything you do that
modifies waves, blips, annotations, or ranges will have those classes
called. It would help to see the code. I personally have never seen
that high of a CPU time,
A gadget for display makes sense, a gadget for editing essentially
skips the entire idea of collaborative editing (WYSIWYG editing makes
it much more difficult to do things collaboratively, as that's a ton
of rendering without canvas). I disagree with WYSIWYG in general
though, as I've used Dreamwe
r ahead, but I went ahead and
> created a Google Code project in case anyone wants to work on/discuss this
> in there. I added each email address in this thread as a project owner.
>
> http://code.google.com/p/wavewebsites/
>
>
>
> On Wed, Nov 18, 2009 at 5:00 PM, Olre
But you
> still have no context to its actual content, like if you want to
> parse
> out any FormElement within the blip. A new mechanism we are working
> is the active gateway API which will allow a robot to make active
> request to wave to retrieve full context of any blip.
>
>
I'm guessing somewhat, but it appears that in your app.yaml, you're
defining wave.png to be the alias for every portion of the site,
overriding both the robot and assets folder. Put the wave.png into the
assets folder (or another folder that you define such as assets) and
try it again. (make sure t
t;
> This is why I think its inheriting from a previous class or some properties of
> the object are being altered. Correct me if I am wrong.
>
> --
> Regards,
> Vikram Dhillon
> On Tuesday 17 November 2009 12:41:30 am Olreich wrote:
>
>
>
> > The Java API has the Blip
Google wave is embedded into Google Contacts, so if there's a way to,
by url, add a contact to Google Contacts, just use that same idea, but
use: yourpseudo...@googlewave.com. And, upon next refresh, the person
that added you, should have you in their contact list. Of course, I
haven't seen any way
They can then untick it and see how
> their content appears...
>
> On Nov 19, 1:00 am, Olreich wrote:
>
>
>
> > Great idea. I think the best method of editing would be to have a HTML/
> > JS/CSS syntax highlighter (Robot). A gadget (essentially just an
> > iframe p
Great idea. I think the best method of editing would be to have a HTML/
JS/CSS syntax highlighter (Robot). A gadget (essentially just an
iframe pointing at a temp-website with the data in it) showing a
rendered version of the web-page based upon the current code. This
will enable the web-page code
This makes me sad I only have a preview account...
On Nov 18, 2:02 pm, qMax wrote:
> missing tags again...
> it could really help ppl to find these waves.
>
> On 19 ноя, 01:00, Austin Chau wrote:
>
>
>
> > This is the public wave for the office hour. Please bring all your
> > Wave API-related q
Each time a robot is added to a wave, a new instance of that Robot is
created. This is persistent across AppEngine deployments. Removing and
re-adding the Robot is the only way to create a new instance based
upon the new deployment.
If you change the version of a Robot, AppEngine will continue to
Ah, my apologies, for some reason, I started talking about the
conversation model and server-server communication lines.
On Nov 18, 2:21 pm, David Nesting wrote:
> On Tue, Nov 17, 2009 at 7:15 PM, Olreich wrote:
> > interface was completely in the terminal). But first, the Wave
>
Try sticking to waves whose blips don't go over about 90. There is
some issue of large waves having serious lag problems. Public waves
are the worst offenders, as they have 300-400 blips often (aka. Wave
Death).
On Nov 18, 10:29 am, qMax wrote:
> I agree.
> I've installed Chrome (linux beta buggy
s to make my life easier :P
>
> Jo-el
>
> On Nov 18, 5:57 am, Matt Richards wrote:
>
>
>
> > On a kind of related note what about intigration with Google Groups
> > (at least being able to pull dicissions from Groups into a Wave)?
>
> > On Tue, Nov 17, 2009
nother thread,
> > > robots can currently do some nasty stuff to waves, I would hate to
> > > give them that kind of power over my account.
>
> > > Also, consider that in the greater "ecosystem" of the wave
> > > architecture, there may be other f
keep the wave in sync
> 5) The wave appears in your wave.google.com inbox
>
> Of course, wave.google.com could be substituted with any server which
> provides mailboxes (waveboxes?) for users.
>
>
>
> On Tue, Nov 17, 2009 at 11:12 AM, Olreich wrote:
> > That is true, the
;
> I don't believe that a robot in this context would have access to any
> waves though, since it wouldn't have a "context" to execute against.
>
> Adam Ness
>
>
>
> On Tue, Nov 17, 2009 at 10:51 AM, Olreich wrote:
> > The API calls that we listed are
Well, there's already a Robot for email, and there's bound to be at
least a gadget and an extension built for it.
On Nov 17, 5:21 am, Raphaël Pinson wrote:
> On Tue, Nov 17, 2009 at 11:16 AM, drone wrote:
> > Guy at Google Developer's Day in Moscow claimed that there are no
> > plans regarding G
ov 17, 10:14 am, Adam Ness wrote:
> Also, wavelet.createWavelet(participants, dataDocumentCallback); in
> java doesn't create a new wave, it only creates a new wavelet inside
> an existing wave.
>
> Adam Ness
>
>
>
> On Tue, Nov 17, 2009 at 12:52 AM, Olreich wrote:
&g
ple code here for Python wave
> > creation:http://wave-samples-gallery.appspot.com/about_app?app_id=60017
>
> > <http://wave-samples-gallery.appspot.com/about_app?app_id=60017>- pamela
>
> > On Mon, Nov 16, 2009 at 1:48 PM, Olreich wrote:
> > > To crea
ke allowing attachments to auto-execute
> themselves when you open an email, and any security expert can tell
> you why that's a bad idea.
>
> Again, I think a third type of "API" would be necessary to support the
> kind of extensions you're talking about here
The problem is that Robots can create a mass of waves, but can't
organize it very well for the user, so the user wouldn't want a robot
to do anything outside of the wave, but rather operate entirely
within. Allowing robots to organize themselves would be expand them
outside of a wave-by-wave basis
The only thing that I can think of is inserting a robot that inserts a
gadget. That's a first guess, as I've had little to do with gadgets as
of yet.
On Nov 16, 11:41 am, Andy wrote:
> I'm trying to build a manifest that will put an item into the new wave
> menu that will create a new wave and in
Lol, I just submitted a feature request for that yesterday xD
http://code.google.com/p/google-wave-resources/issues/detail?id=462
On Nov 17, 1:03 am, jhb wrote:
> Thanks for the discussion. I created a feature request and added some
> of the ideas discussed.
>
> http://code.google.com/p/google-
The Java API has the Blip.getParent() function. When this is called,
the returned blip does not have the ability to "getParent()" again it
seems. or, at the least, one cannot continue walking without hitting
null on the second trial. The same rings true when using
"isParentAvailable()" to check for
If robots get access to folders and/or tags, that organizes the inbox
pretty easily. Also, both Java and Python have persistence models, so
an extra wave for Data is unnecessary. One could also use a bunch of
Data Documents for the data as well.
On Nov 15, 9:44 pm, Jo hyphen el wrote:
> I like th
Python is extremely limited in how it interacts with waves (notably,
it can't delete blips).
On Nov 15, 11:26 pm, iJames wrote:
> Oh! The Java API has more functionality?
>
> I'm really liking playing in Python too!
>
> On Nov 15, 6:54 pm, Olreich wrote:
>
>
>
To create a wave via programming in Java:
wavelet.createWavelet(participants, dataDocument);
wavelet -> A Wavelet object.
participants -> A List of the participants that you want to
include (format of name: usern...@googlewave.com)
dataDocument -> A String referring to a named dataDocument where
When I changed the version on my robot, I found that appEngine kept
serving v1 of the application. To fix this, I went into the "Versions"
panel on the dashboard (under the Administration header) and made v2
my default version. Then removed and re-added the bot, and everything
was updated well.
--
As a temporary work-around for Wave slow-downs and as a method of
education, I've decided to build a quick little robot that will delete
the second blip (the one after the root blip) when the blip count is
greater than 20 and then archives the data from that blip into the end
of the root blip. This
Python, but currently Java is the only worthy candidate because of the
fully fleshed functionality. I much prefer programming python though,
it's so much smoother.
--
You received this message because you are subscribed to the Google Groups
"Google Wave API" group.
To post to this group, send em
62 matches
Mail list logo
