You have no idea of the quality of the data set. You have no controls. You have no way of knowing how the 353 entries are biased.
You make the claim that it’s just math and the math can’t be wrong. But that is a lie, and from the levels of education you have claimed, you must know it is a lie. And you have been called out on this a number of times…so it’s intentional.
This is exactly the way statistics are used to tell lies. The façade of mathematical precision obscures the shaky foundation of the data. By continuing to extrapolate what you believe is the current observed production rate to a completion time (or more precisely stating a required linear production rate to complete on time) continues to ignore that production lines in startup mode are not linear.
What the F**k. Maths has never been proved to be wrong ever. Even the universe has to follow it.
Where am I extrapolating?
What we can see is the rate at which the new emails and deliveries are sweeping through the order days. It doesn’t really matter how big or small the data set is, or if it is biased in some way, unless you think they are really on day 5 but all the day 5 people omit to fill in the spreadsheet. Or do you think GF are avoiding sending machines to people on the spreadsheet, so that it under represents the machines dispatched?
And since there are N people on the spread sheet with no GF and D days left we can say we need to see an average of N / D deliveries per day on the sheet to be on target. Yes it might start less than that and ramp up but each day it is less it has to ramp up more because D gets smaller faster than N. And it isn’t anything that depends on the quality of the sample.
Going to pipe in and remind people of the value of civility.
Statistics is a fun little beast where if you know about the pitfalls, they are painfully obvious. And if you do NOT know about them, they are impossible to intuit. So do NOT make presumptions about what anyone else should or should not know about statistics. I would venture I can school you all, but I would still claim that I know essentially nothing about the topic.
Math doesn’t lie, but it only answers the formula, not the questions which led to the development of the formula. If you value the spreadsheet, fantastic. If you do not value it, okie-dokie.
No point in pushing even this close to flaming one another.
So tell me how the data in the spreadsheet is not accurately showing the progression through the order days.
It isn’t garbage data. That is very insulting to the people that have entered their data accurately. I expect there might be the odd typo and the odd person that gets a machine and forgets to update it, but plenty of people do update it so we can see how quickly GF are getting through the order days.
I’m not going to continue to discuss this. The individual datum are likely fine. But taking the data set as a model for the whole is unfounded. That’s what I’m applying GIGO to.
Well just calling it GIGO, or calling statistics lies, etc, without any evidence is fallacious. In what way do you think the sample isn’t random? Since people self select by adding their own data I don’t see how that can generate a bias regarding when they will get their machine unless GF base their deliveries on it.
If anything, since it has a few regulars that got their machines early, the total probably over represents the total machines sent out. I.e. 10% of people on the spreadsheet have got a machine but I doubt 1000 machines have been sent out as they would have got further through the orders if they had.
Regardless of any non-random bias it shows the progression through the orders, which is what everybody wants to know. I can’t see how it can mislead on that. If there is any big ramp up we will see it straight away because they should sweep through each day’s orders in two or three days of production to meet the schedule.
I see the spreadsheet as being a good general indicator of where Glowforge is at in terms of order day versus shipping. In other words, if several people who ordered on day 5 get their emails, we have a general idea that Glowforge is shipping to those who ordered on day 5. This has value for someone who ordered on that same day, because they’ll know with at least a small degree of certainty that their order will be shipping in the near future. Trying to determine, with any reasonable degree of accuracy, exactly when any one person will receive their Glowforge, or how many are being produced/shipped per day, is still just a guessing game. Without knowing how many orders were placed each day, how many are produced each day, how many are shipped each day, how many will be produced and shipped each day going forward as production presumably ramps up, we really don’t have the ability to predict whether the goal of delivering everything by the promised date is achievable. So, the value of the spreadsheet, at least to me, is that it gives a general idea of the speed at which Glowforge is fulfilling orders. Based on my order date, I still have a long wait ahead of me. I think I’m starting to ramble now, so I’ll stop typing. I hope my point was made without fueling either side of the argument. Now I’ll continue to wait for my Glowforge, and the pizza I ordered an hour ago that should’ve been here by now.
To say that seeing this spreadsheet is upsetting is an understatement. I ordered my Glowforge on October 9, 2015 and I haven’t received an email yet. How can so many more people who ordered after me have received theirs? Maker faire shouldn’t put you ahead of those of us who have been waiting longer. Whatever happened to right and wrong? There is a right way and a wrong way of doing something and it looks like they are picking the wrong way. I have been waiting patiently coming here reading about everyone else receiving their glowforge and being happy for them because I thought they were all day one people.
Don’t even see your name on the tracking spreadsheet … so if that is your source of information on deliveries --> why don’t you participate?
The 11 or 12 (according to the spreadsheet) with order dates after 9 Oct 15 are all there for explainable reasons … received PRU’s, have added immeasurably to the development of Glowforge thru their testing efforts, are attending Maker Fairs where there talent will bring to good light the remarkable how makerspace and easy to use laser that we know as the Glowforge.
Right or wrong … the inevitable delays WE HAVE ALL WAITED THROUGH TOGETHER have, and will, yield a remarkable product that we said to @dan and crew we wanted. Yes, many of us are and remain frustrated.
Perhaps the best thing you can do is befriend an owner and ask for guest permission and log onto app.glowforge.com and see what the design experience is like so you can be up and running as soon as the 70 pound beast arrives. Being able to do this whet my appetite … but more importantly has made me a tad bit more patient.
