Use this space to cheer the creator along, ask questions, and talk to your fellow backers. Please remember to be respectful and considerate. Thanks!
You are right, I have seen a couple of projects try that, the analysis would be a nightmare!
Mind you one KS did it so that negative comments would get buried amongst the votes!
It puts me in mind of a favourite maxim of an old boss during our creative design sessions: "Never criticise unless you can offer a better alternative".
The "Coolest Cooler" Kickstarter did a color choice survey by just asking a question in an Update instead of using a survey site, and the result was several thousand Kickstarter comments all voting for colors. It was a mess. I much prefer the use of a survey site.
Whoops! I missed a critical word.
These tools do not NORMALLY store the email addresses with the survey host.
Some do, if you use a webservice rather than a downloaded prog. Such data are normally stored in encrypted form.
I totally share your desire to keep private information private, but you seem to be getting worked up about a normal survey practice used by many websites and KS projects.
Companies often use third party survey tools, as they are much better than inventing an own one and getting it wrong. I used one (Websurveyor) when I ran a consumer panel for my company, so I know how they work.
These tools do not store the email addresses with the survey host, they normally refer to the list on the customer's system. I have answered many such surveys, using an email address that I use solely for Kickstarter and nothing else. I have never had spam emails arriving, so I guess they are trustworthy.
The great advantage of using such survey tools, is that they offer easy to use conditional and branching surveys, cope with multiple response from the same customer, and offer built in analysis tools.Some even allow the overall results to be published back to the web as charts or graphs.
It sounds to me like with the terms you laid out the creator would be allowed to sell the email address to further their cause. Hey the extra money could help right? That's bull crap..
I may sound crazy but I do my best to keep my information private. I'll make sure to email KS about this.
James - the conspiracy buster!
He put that to bed pretty well.
@Tyler Smith, Survey monkey didn't send the email (at least the one I got). My email was from the Scio team through Mailchimp (an email management platform that doesn't keep the emails) and I clicked a link that lead to the survey.
Even though it doesn't appear that SurveyMonkey got anyone's email, their receiving it appears to be within the specified agreements, but then again I am not a lawyer.
@Tyler Smith, they offer a free version for a reason, right? They have to make money some how...
@bobbunting. The creator had to enter my email address in their system to send over the survey. I did not ok my private information(email address) to be sent to a third party. I've been doing some reading about surveymonkey. They don't sound very reputable to me.
I vote for the black button anyway!
Why didn't you notify us through the KS updates, I would not have know about the survey because I receive too much junk email
There should be two votes one for "your" color choices and the second for what the majority wants(write in), then you can see if it is very popular and have a second survey if it is an overwhelmingly a popular color.
For those who only monitor the KS updates and notifications, you should post a survey notice was emailed out on KS
I am just confused because I thought we got to pick the colour that we wanted. Not have to settle for some group choice.
I like Heather's idea of a second poll on the top 3.
If we vote up the Metallic cover shell...we get a black button...and a black device. The cover isn't an important aspect.
I will vote for dark grey, most other devices I have are black or silvery grey. The brighter colours may attract too much attention when I am checking out the olive oil quality in a shop :)
Actually any colour will be fine, sooner rather than later.
The color survey this large should be done twice I believe. Once this is narrowed down to the top 3, then we should all vote again - if you want a 'true' opinion of everyone for the final colour.
It is surveymonkey, probably one of most used and reputable online survey companies. You DO NOT have to log in, or sign up, or enter an email address. Sometimes I think a lot of KS backers, just crawled out from under a rock.
I LOVE the BLACK BUTTON, why only an option with the Metallic ? It should be an option with the other darker colors too.
...and there is the drawback to using only one email address. Grab a few more addresses and not worry about it.
Tyler... you have the option not to make a chioce..or choose one of the other colors that are not painted.
Also who's to say this third party survey company won't use our email addresses later? I don't like this at all. I never authorized my email address be used in a third party site.
This is shady business.
Why is color even a question? Who the heck cares. Just get us out LATE OVERDUE product to us ASAP.
Also why would you run a survey through any company other than KS? I'm sure it went into my spam box. I've not seen it yet......
Also on a side note, painting sucks. 1-2 weeks I. My pocket and the pain will be look like crap.. How about not painting it?
What color indeed... I like the Metallic...but it will look ratty with wear after a little while..I also like the black button of the metallic version. I could always repaint later. So, maybe a good device color with a DIY exterior paint job.
Why only the different button color on the metallic version?
Delivery dates in Kickstarter campaigns are estimates. When developing a new product, there are usually unforeseen issues that lead to delays. It's all part of the process.
Consumer Physics posted a link to a progress timeline here:
I'm past the point of wanting a refund... in fact I am considering my money gone with zero return. I feel like they have enough partner support by now to refund backers, but that doesn't matter... Every time I see "we won an award" I think, how did they win an award for a non existent product? if it is finished and winning awards then where is mine? Kickstarter has been nothing but good ideas with great disappointment.
semi unrelated, but I am working through this book to get up to snuff with basic chemistry:
It's an awesome book
Thanks Roger, when I get my SCiO, I will be looking to you for help! :P
Thank you, James and Roger. That's kind of what I thought/expected, but wasn't sure. I did not do well in chemistry... :-) Love techy toys, though!
This link shows how different methods were used to estimate fruit sugar content (NIR included).
The author has a similar article on sugar in drinks.
CP have said that "standard" SCiO users will be able to make scans that will be part of reference sets, although they have not explained exactly what the validation process will be.
I would expect that the developer / researcher interested in that subset will screen such scans and choose to include or exclude them.
Obviously a scan on an unknown material will have less value.
There is much detail to be explained yet.
@Laura Ling - Everything I've read suggests that for now you need to have the developer SDK at least to upload new models, but an existing app may update it's model based on scans, such as if the formulation of advil for their pill app is slightly different in the US vs UK vs Israel etc.
I think SCCiO team should update the picture here on kickstarter like their homepage which uses the new design case as all leaving the old one does is show you've hired krusty the clown to manage marketing and a makerbot hackerspace session to redesign it.
I apologize in advance if this has been answered already... ;-) Will the scans we take be added to the(a) database? Or only scans from developers or scientists who can provide information on what was scanned?
I hope that is the case.
I think you misunderstand the model, and how it interprets a new scan.
I can equally well be barking up the wrong tree, but I did have several interactions with CP on this subject.
Let's start with fat percent in cheese as an example, and let's say that we trust manufacturer's labels as a start. Probably better than my kitchen lab).
When we scan a lot of samples with higher and lower fat%; and allocate fat% to each it will create a the basis for estimation of an unknown sample. It will not look for nearest match, but best FIT within the whole sample set. There may be 10 reference samples all claiming 40% fat, but all will have different absorbance values, another 15 samples may claim 25% fat, and a further 12 samples claim 55% fat. Then, when a test sample with 35% fat is scanned it will use all reference scan results to estimate the answer. Much better than Google could do.
Indeed with time, some of the reference samples may rejected as outliers because they do not fit the general pattern.
I really would like CP to explain this more, especially if I am wrong!
Apparently I did not do my homework enough either. I expected that this would estimate the values based on a database, not pull the exact data from the database. People were accusing SCiO of just identifying the product and doing a google search of it. To me they are doing just that, but not using google, instead their own search data base, with no value estimation. In fact, google would have been better as it would probably include more data. I don't have a lab, so any info that I upload will probably be from google (or the label), so why not just cut the middle man and take the data right from a google search?
No Worries Michael - KS comments keep us all entertained when it is too cold to work in the garden :)
My apologies for not having done my "homework" seems to be in order.
Thanks Lothar, maybe in a couple of years the SCiO Mk V will cover UV and IR both :)
I am continually surprised that backers keep asking if SCiO will detect substances that exist at microgram per gram levels. It is in the FAQs (as you kindly posted) and discussed ad nauseum in ever so many comments. I guess it is easier to ask than do the research!
since CP stated that some substances can be detected down to ,1% or less, Vitamin C might be characteristic enough to be detected but UV not near IR !
I see you also picked up on CP's comments about seasonal and regional variations for fruit and vegetables.
In addition, I wonder how much composition variation will be caused by degree of ripeness, and whether a Granny Smith apple will scan differently to a Gala. I think there will need to be plenty of classifications / metatags to get clean results.
But mostly I wonder how the values will be verified.
Please remember that the minimum detection limit is around 1% (depending on the material).
I strongly suspect that cyanide, anthocyanine, and vit C are present in much smaller concentrations.
My reference book says raw tomatoes have about 20 mg vit C per 100g ... so 50 times too small to get noticed.
(The Composition of Foods by A.A.Paul and D.A.T. Southgate 4th edition)
thankyou for your quick response, then based on that, that too holds true for analyzing ie Californian wine, and determining cyanide levels? (its currently a thing)
@Michael, someone can correct me if I am wrong, but that is only possible if the exact verity for a particular year of tomatos has been measured for vit c and anthocyanins with a large enough sample size to provide a resonable average. Then you will get that average as a result of the scan. Plus an app for that needs to be made.
Is it possible to determine levels of Vitamin C & Anthocyanins present in a tomato with the SCiO?
I was also disappointed with the redesigned case, and wrote a PM to CP about it. Their reply gave good technical reasons for the change, but I still miss the sleekness of the original, and the lanyard attachment point.
It would be great if someone with the appropriate mechanical skills could design such a casing around the final SCiO internal components.
I do however think that it is too late in the project to push CP for a redesign before shipment, otherwise the SciO will arrive as Xmas presents!
BTW.: Anyone capable of product design is invited to make his own case and provide it at a 3D printing website to be used by us backers. I'll try to do so if I have the SCiO here in real. It may be only a cover or partly replacement, everything to make its look more 21st century.
you can scan pills without the holder of cause, but since a pill might be small to have a defined surrounding you should use it to get constant results.
@all, @ Consumer Physics,
although a design change might have practical reasons,
NOBODY LIKES THIS LOOK!
(except some notorious yeasayers)
I'm also in the category of people that would also like to get the raw scan data from my own scans. I don't need to be able to access it from your database, but it would be VERY cool if we could have the ability to store and tag our own scans on our mobile devices. Admittedly I haven't been following comments closely for the last few months -- is this something that you will let us do?
I suggest that if a scanned item is not identified because it is not present in the database, the app should show the spectrum info and measurements for the user's own research. Also, the app can show the 3 closest items in the database, and by clicking on one of the items, it should show the similarities between the scanned spectrum and the item's spectrum.
Is the pill holder required to accurately scan and identify a pill?