W3C home > Mailing lists > Public > public-device-apis@w3.org > December 2010

Re: Comments on http://www.w3.org/TR/2010/WD-system-info-api-20100202/

From: Philip Gladstone <pgladstone@cisco.com>
Date: Wed, 15 Dec 2010 18:17:28 -0500
Message-ID: <4D094C88.5050809@cisco.com>
To: Antoine Sanchez <ckkoshi@gmail.com>
CC: public-device-apis@w3.org
Thanks for this. With regard to the level, you could not report a number
less than 0. However, most platforms start a shutdown process at some
poinrt, and it seemed that that point was defined as zero. Given that
the shutdown process can take some time (and power), what value do you
want to report during the shutdown process? I think that the same is
true for the maximal value. Most rechargeable batteries lose capacity
over their life. Does this mean that, once the battery has lost 50% of
its capacity, the value never reaches above 0.5? Or is it scaled on each
charge cycle -- so that 1 means that the battery cannot be charged any
more? NiMH batteries lose capacity when the temperature drops below
freezing. Does this factor in? My main point is that, if applications
are going to do sensible things with these values, then they need to
understand what the values represent.

Is it possible to charge without external power? It depends on your
definition of internal stored energy. With the rise of ambient energy
harvesting technology, it doesn't seem unlikely energy could be
recovered from a spinning disk as it is spun down (think what your
Toyota Prius does today when you apply your foot to the brake.....)  You
may think that this is a corner case today, but I'm sure that a lot of
people said the same about recovering energy during braking in the early
days.

Philip

On 15-Dec-10 17:35, Antoine Sanchez wrote:
> Hello,
>
> I'll just give an answer about 4.4.
>
> About level, I don't see the point why there's a negative value, 0 is
> enough to say there's no more power, can you explain?
>
> I think if isExternal is false, there's no way that isCharging can be
> nothing else than false.
> If there's a solar charging, or even micro-wave charging, power source
> will be external, not from battery.
>
> Thanks,
> /toni
>
> On Wed, Dec 15, 2010 at 10:58 PM, Philip Gladstone <pgladstone@cisco.com> wrote:
>> Hi
>>
>> I note the the StorageUnit property does not include flash memory. Since
>> this is the most common sort of memory in phones (and probably in next years
>> laptops in the form of SSDs), it seems a pity to use TYPE_UNKNOWN.
>>
>> I also have a philosophical objection to TYPE_UNKNOWN as it occurs in many
>> places in this document. In some cases the type may really be unknown, but
>> in most cases it is really TYPE_OTHER (such as the flash example). You may
>> want to consider whether to add a TYPE_OTHER to each (effective)
>> enumeration, or whether to define an extension mechanism to allow new
>> technology to be adequately represented.
>>
>> In section 4.4 Power, it is clear that a very simple model of power is being
>> assumed. It isn't obvious how you would fit an integrated solar charger for
>> a cellphone (or e-book reader) into this model. When the sun is shining, is
>> 'isExternal' true or false? Since the values of isCharging and isExternal
>> have specific relationships, this gets complicated very quickly. If the
>> solar charger was external, would this make a difference? I have seen
>> laptops with fuel cells and batteries, and even one with a hand-crank.
>>
>> I would define the terms as follows;
>>
>> level of type float, readonly, nullable Specifies how much of the internal
>> energy source remains, scaled as follows. A value of 0 means that the stored
>> energy level is at a point where the system will enter shutdown mode, and 1
>> indicates that the system's stored energy level is maximal. Negative values
>> are only likely during the system shutdown process. Any threshold parameter
>> used in a watch function to monitor this property applies to this attribute.
>> No exceptions.
>> timeRemaining of type unsigned long, readonly, nullable Represents the
>> estimated time remaining in seconds before the system enters shutdown mode,
>> based on net current power consumption (which may be averaged over a short
>> period of time) and how much stored energy remains. If this value is null,
>> this means that there is essentially infinite time remaining.
>> No exceptions.
>> isExternal of type boolean, readonly If true the device is currently
>> receiving any power from an external source. If false the device is not
>> receiving power from an external source.
>> No exceptions.
>> isCharging of type boolean, readonly Indicates whether the internal stored
>> energy  level is currently increasing. If isExternal is false, this value is
>> likely to be false, meaning that the system is consuming net positive power,
>> and the stored energy source is therefore depleting.
>> No exceptions.
>>
>>
>> In the case of the section 4.7 Network, there seems to be an assumption that
>> there is only a single connection to the network. This isn't the case for
>> nearly all systems today. Also, with the advent of v6, each interface will
>> have multiple ip addresses (probably at least one v4 address and probably at
>> least two v6 addresses). This doesn't seem obviously representable in this
>> schema.  Even if the interface was changed to return multiple Network
>> property objects, it isn't clear whether it will return interfaces that are
>> down. For example, would it return a TYPE_IEEE802_11 object if the Wifi was
>> not currently associated to an access point (or a member of an adhoc
>> network)? I suggest that having a way to enumerate the down interfaces will
>> be useful. Further, there are logical interfaces such as VPNs or general
>> tunnels as might be used in an I-WLAN implementation. It is not clear how to
>> represent these. I also note that an application cannot tell what interface
>> will be used for any particular network operation as it does not have access
>> to the routing table. A use case might be to delay downloading large images
>> while connected over GSM (and only show thumbnails) but show the full images
>> if connected over 802_11. The problem is that the app cannot tell whether
>> the connection to a particular data source will use GSM or 802_11. For
>> example, if the 802_11 interface is associated to a public access point,
>> there probably needs to be some sort of captive portal remediation before
>> connectivity is established to the Internet. If the user chooses not to pay,
>> then the connection will remain up at layer 2, but not able to communicate
>> with the Internet.
>>
>> Thanks for reading this far...
>>
>> Philip
>>
>> --
>> Philip Gladstone
>> Distinguished Engineer
>> Product Development
>> pgladstone@cisco.com
>> Phone: +1 978-ZEN-TOAD (+1 978 936 8623)
>> Google: +1 978 800 1010
>> Ham radio: N1DQ
>> Blog: http://wwwin-blogs.cisco.com/pgladsto/
>>
>> Cisco.com - http://www.cisco.com
>>
>> This email may contain confidential and privileged material for the sole use
>> of the intended recipient. Any review, use, distribution or disclosure by
>> others is strictly prohibited. If you are not the intended recipient (or
>> authorized to receive for the recipient), please contact the sender by reply
>> email and delete all copies of this message.
>>
>> For corporate legal information go to:
>> http://www.cisco.com/web/about/doing_business/legal/cri/index.html
>>

-- 
Philip Gladstone
Distinguished Engineer
Product Development
pgladstone@cisco.com
Phone: +1 978-ZEN-TOAD (+1 978 936 8623)
Google: +1 978 800 1010
Ham radio: N1DQ
Blog: http://wwwin-blogs.cisco.com/pgladsto/

Cisco.com - http://www.cisco.com

This email may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message.

For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/index.html
Received on Wednesday, 15 December 2010 23:18:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:14:15 GMT