The 80 20, 20 80 Rule

The title pretty much sums up my situation. The first 80% of the project takes 20% of our time, but the last 20% of the project takes 80% of our time. We have flattened all of the “easier” bugs and issues, but now the “big” ones remain. However, the question of how “big” these issues really are, has begun to arise.

No, we are not trying to be lazy and shirk away from hard work. We have simply realized that some projected goals from the beginning of our progress are not practical.

 

The current bugs we are trying to flatten out right now are not really bugs that take away from the purpose and usability of the SuN program. In fact, they don’t really matter at all, and the program runs perfectly fine. Because the Space Suit Case is intended to be used by teachers and museums -not researchers, it is not really vital to have every single picture of the sun for every single day.

Because the page is loaded with over 100 images (days), missing a few images will not take away from the intended purpose of the SuN. Keep in mind that the images (days) that are skipped are scattered throughout the 100+ images, not all at once or back to back). The code will always retrieve 100+ images (unless the SDO shuts down or something), and so the user will always have enough images to observe and navigate a smooth time-lapse of the Sun.

 

Another “big” issue, is that the images in one particular wavelength set are smaller than the images of the other wavelength sets. This results in the images of the SuN in that wavelength set not filling the entire sphere. The images in this wavelength set are the same size (1024x1024px) as the other images in the other wavelength sets, but in them the Sun is smaller and there is more black space. So if we try to enlarge the images in that set, it will actually unalign the other images from the other set so that they are not centered on the sphere. So, in order to keep this wavelength set, we would have to 1. enlarge all of the images, and then 2. crop all of them before loading them. This doesn’t seem impossible, but it does seem kind of impractical

Coding a unique loading process for the images in this set would be doable, but the problem is that it would slow down the entire program. Our code already isn’t the most wonderfully optimized, so making the code run a completely separate loading process for these images that involves cutting and resizing the images, would not be very welcome. In fact, it may take away from the user experience by extending loading times by a substantial amount. We already have 4 perfectly function wavelength sets, and this 5th one very closely resembles the other 4. Removing the wavelength entirely seems to be the best option. Again, we are not trying to avoid work, it just seems that even if we do successfully get it to work, it will bring undesirable side-effects that are impractical for the purpose of the project…

I would like to thank Maryam for her selfless donation of her time and effort in order to explain computer networking to me and help us approach our issues. While I don’t know if I will be using much of what you taught me for my project anymore, you still inspired me to learn more about how computer networks and the web works. It’s something that I have always taken for granted, like electricity. Also, special thanks to Bob for always taking the time to make sure that I’m okay and that I have everything I need.

Advertisements