Project Proposal - University of Illinois - Engineering Wiki

quarterceladonΚινητά – Ασύρματες Τεχνολογίες

10 Δεκ 2013 (πριν από 3 χρόνια και 10 μήνες)

112 εμφανίσεις



[Type text]




{String.Empty()}

Project Proposal

Aaron Dai

Alex Sha

Janus Yuan

John Fritzel

Michal Krupa

Xu Gao





2

Introduction & Idea Brainstorm


The first step in coming up with our project was generating a “tag cloud” of
words from which we could deduce
a tangible idea. We attempted to connect as
many concepts as possible in order to overlap several sectors of relevant subjects
that would produce a well
-
rounded idea. We used lines to connect them.




Our ideas were rated based on
innovativeness
,

i
nterest
,
significance
, and
feasibility
. We assigned point values to the above eight generated ideas, which we
were able to scale down to five using this rubric. The resulting ideas are below
:





3

After careful consideration, in addition to having a much great
er point value
in comparison to our other potential applications, we figured the interview Q&A app
would be the most relatable to a college audience. Additionally, it was something
that we could all easily become engaged in as college students ourselves wh
o have
had trouble finding relevant and useful interview questions.


Addressing the Problem


First impressions are very important. The signup and login screen is the first
stimuli a new user will be introduced to. This screen is designed in the spirit of

simplicity and clarity. It achieves simplicity by only including two buttons: Signup
and Login. The signup button is instantly recognizable, because of its size, and its
color, which match
es

the color of our application’s logo. This screen achieves cl
arity
through the Ill
ini “I” in our “Interview” logo,
which creates an association between
the University of Illinois and our application. This association will make our
application feel relevant for Illini students. Beneath the logo, our mot
t
o “Ace your
next job interview”, will instantly allow a user to realize the purpose of our
application. This purpose is expanded
upon
during the actual signup process with
the message “Get instant access to hundreds of real interview questions.” The
important keywor
ds in this sentence, “hundreds” and “real”, are highlighted for
additional emphasis to further persuade new users.


Upon signing up, it is important for users to be able to quickly find
companies


which interest them. Two navigational features are prese
nt: a search
bar which allows a user to do this manually, and underneath this, a series of
company logos which are recommended on the basis of the user’s major. It should
be clear to a user that they can speed this process up by importing their LinkedIn
p
rofile. This option is rectangular rather than square as to not confuse this option
as a “recommendation” rather than an additional feature.


It is important to allow users to become accustomed to features by providing
cohesion

throughout different area
s of the application. The first time user will
select
companies
that

interest them, and they will later return to a similar page
when searching for updates about these companies. Companies
that

have new
activity will be assigned with a tiny yet visible r
ed bubble containing the number of
new updates.


As

interview questions and answers are the driving forces of the application,
a method of organizing them is crucial. Questions are displayed with their parent
company logo to allow a user to determine if t
he question is relevant to them, as
well as providing organization. The answers to a specific interview question are
organized by popularity, which allows a user to quickly sift through unhelpful
answers. The popularity of an individual answer or questio
n is determined by the
amount of “likes” it receives, which is visible as a numerical value to the right of
each question or answer box.


4

A frequent user will be interested in a concise yet detailed display of their
achievements and progress in the applic
ations user ranking syst
em. A single
“achievement” bar,
which fills depending upon the user

s progress
,

achieves this
clarity. Above this bar, three additional bars provide details about a user’s progress
in the categories “Companies followed”, “Comments

made”, and “Interviewers
rated.” These bars are smaller as to cue a user that they are a part of the larger
“Achievement” bar and do not stand alone. Beneath all of the bars, our ranking
system provides a visual “badge” that represents a summary of a us
er’s current
ranking. This allows a user to feel rewarded for their achievements with a “prize”,
rather than raw statistical data.


Analysis of Existing Solutions


The solutions that currently address the recruitment feedback loop are
minimal and provide
little insight into the actual interview process. Glassdoor keeps
track of interview questions and reflections submitted by users. They store the
method of application (online, referral, campus recruiting, etc.) as well as allow you
to rate the interview e
xperience, but Glassdoor lacks key insight metrics.


Due to the structure of the feedback system, you are limited to a positive,
neutral, or negative experience ratings and providing information on whether or not
you were offered the position. When a user

rates their interview, the positive and
neutral experiences are associated mostly with receiving a job offer, accepting or
rejecting it, and there are mostly negative responses for interviews that resulted in
no offer, as well as neutral overlap. The numb
er of neutral responses for “no offer”
as well as “accepted” and “rejected” offers are about equal, so this metric provides
little useful information. Most of the reviews left are for rejected or no offer
responses. This shows a bias towards leaving respon
ses in the case that you were
unsatisfied with an interview, skewing the content and value of the service, and
there is no system that accurately evaluates the usefulness of negative content posts
in comparison to the positive content posts.


There is no t
racking based on which step in the interview process an
applicant has passed to. By allowing users to enter questions and experiences based
on the current step in the interview process, (first phone interview, first in person
interview, etc.) we get a much

richer databank to analyze content from. We can use
information like satisfaction based on the interview step to appropriately weigh the
satisfaction level for a certain position
-
for example, if the first two rounds went very
well and the third was neutra
l, we can weigh the users satisfaction more heavily
towards a positive experience and keep track of the user as an individual with
possibly higher interview expectations. Additionally, this provides us a way to sort
out negative reviews more fairly and aut
omate the alleviation of bias within
displayed content. This “weighted rating” is something we can use probability for, in
tandem with an applicants rating history, to scale a more balanced rating among
various members in order to get a relevant “total rat
ing” per position.


5

About.com and other general sites provide information on interview
questions and the interview process, but these contain articles and tips rather than
user generated content reviews. The common element in both the static content
sites a
nd user
-
generated sites is that they allow visitors to remain anonymous. This
is a functionality we plan to replicate in order to promote honest responses without
the risk of exposure to potential recruiters or companies. No existing site caters
exclusivel
y to college recruits, which limits the ability to sort content. Our vertical
search engine and scoped community provides much more relevant results, and
allows us to target an 18
-
28 age group demographic, grasping a niche market that
will continue to rapi
dly expand as the collegiate space grows. Additionally, in
expanding the solution to a larger user base, we already have a reputation among a
young demographic that would continue to see us as a professional information
source.


Proposed Solution &
Benefits, User Base & Product Value


We are making a platform for students from U of I to efficiently share highly
related interview questions with each other. If you try to search on the Apple App
Store, you won’t be able to find any app that has this
function. Thus, we will have
two types of competitors, interview preparation apps and int
erview questions
sharing websites
. For interview preparation apps, they are not real time interview
questions; they only have an interview questions database, which is

going to be out
-
of
-
date. For interview questions sharing website, you might be able to find some
interview questions with your company. However, someone in a different city posts
those questions with a completely different department in the company. Ninet
y
percent of the time, you will get a completely different type of question in your
interview. Unlike us, those interview questions sharing websites are not focused on
specific position, specific location, specific interview
round, or specific department.
The f
unny thing here is that larger companies usually ask candidates the same type
of questions. Interviewers are usually full
-
time employees at the company, and they
are not professional interviewers. Therefore, they won’t spend too much time
developing n
ew, creative interview questions for interviewees, especially in larger
firms. For example, Microsoft will always ask questions about Linked Li
st data
structure in the first round, Jump T
rading usually like
s

to ask

about

cache, and
Goldman Sachs always ask
s

about object
-
oriented programming. Now let’s take a
look at the screenshot of the searching result from glassdoor.com below. It is either
“I don’t remember” or a completely unrelated question. Our app can bring user the
real
-
time and highly focused inte
rview question for that year, certain position, and
even specific interview round. Thus, we will have
a
unique functionality

that we
provide

users as well as
an
advantage over competitors.







6



There are
a
couple useful values that users can gain fr
om our app. First,
interviews are scheduled throughout the day, so students who have the afternoon
interview can benefit from the students’ post from the morning section by looking
up answers for questions or even understanding what type of person their
in
terviewer is. Second, if there is a very creative interview question, it is very easy to
drive an active discussion after the interview. It is like solving a fun puzzle.
A u
ser
can build their reputation through this and at the same time feel a sense of
ac
complishment. Third, interviewees can rate the difficulty of each interviewer. If
the company allows students to sign up for their interview, this will be very
valuable. In addition, students will know whose interview they should spend most
of the time pre
paring for. Finally, users can know if the companies

have

already
decide
d who is going into the next
round when they are still on the company’s
waiting
list. This will help students

get a better understanding of where they are in
the interview process with
out sending awkward emails to recruiters or
interviewers.

At the same time, reciprocity will drive people to share their interview
questions after they get the benefits from others. In addition, we will establish a
reputation system with badges and “like
s” on questions or answers to drive users to
come back to the site and contribute more. We will also use function
s

like share to
spread the app and attract new users. Since we will be the first
-
mover, we can also
get our competitive advantage through large
r existing question and user database.











7

Purpose &
Practical
Application


The
interface following the login
screens

will allow users to choose the
companies they

are

interested in. Inside
this interface, there are many compa
ny

logos. When
a user

see
s

one
, they will click
on
it
to select

that company
. Once a logo is
clicked, there will be a check mark on the
logo to indicate that it has been successfully
select
ed
. There is also a search bar and
search button for
the
user to type in the
company
name and search for it. This
selection interface is similar to many other
app
s
. For instance,

Twitter

uses
a
similar
interface to allow

the

user to choose whom
to follow
,

and
P
iazza
C
areer also uses
a
similar interface to pick the companies
users are inter
ested with. Therefore, our
users will have intuition
on how this
interface works
. At the bottom of this
interfa
ce, there are "import profile [from
LinkedIn]
" and "next" button
s
. Users who
do
not know what the "import profile
"
button
does

will click on it a
nd the app will
go to a page asking
for the
users'
LinkedI
n
credential
s

and tell users
that
we will grab the companies
they follow
on
LinkedI
n
and select them as the companies users
are “
interested

in” following

inside of our
app. Similar
apps are doing th
is usin
g Facebook/Twitter account authentication

to
give users a faster sign
-
up experience, s
o this feature should not be a surprise to our
users.



Users' profiles contain a
page that

shows their achievement.

When
a user

see
s
this page and the percentage
s for each bar, users will know how they
compare to
other
users. Metrics such

as achievements and badges
are not new to any users.
Again, there are so many other apps doing this to make users more active.



Once there are new interview questions posted to
a company's
page
, there
will be a
notification

count for the number of unread new posts
.
The
notifications

are similar to unread Facebook notices/messages. Therefore, when users see the
notification

on
a

company logo, users will know to click to see the ne
w posts. After
users enter a specific company
page
, the next interface provides users

the

option to
choose which positions they interested in.
Afterwards,

users will
be directed to a

page
that

contains all the top viewed questions related to the selected p
osition. The
three dots under "top viewed questions" tell the users that they can

swipe to the
side to access additional question posts
. At the bottom of this page, there is a button

8

for posting new questions. The page to create a new question contains a t
ext box
which says "Type question here…". Users will know
this

is the place to type in
a
new
question.
Below

the text box
tags

can be selected and the active tag will be
highlighted. In
this

way, users will know to
click

on the darker
-
colored tags if they
want to select a different tag. To submit the question, users will click the "submit"
button at the very bottom.



Inside the interface
that

shows a particular question and the
answers/responses for the question, there
is a

like


button (
similar to Facebook’s
)
for users to

like


the question or response.
There is also the ability to bookmark

each question/response. Once
a user

clicks on the bookmark or the like button, it
will change color to indicate
to the user

that
it is bookmarked or liked.
The
bookmarked items will be displayed inside the Watch List interface
.



At the bottom of the interface,
there is a navigati
on bar. By selecting
the profile silhouette
,
a user

will
know
that
it is

related to
their

profile.
The
button with
a
magnifying glass directs
users to the company search page.


Implementation Approach &
Feasibility


To ensure the quality of our
application and
on time delivery,

we
have
decide
d

to implement the
application using JavaScript
for cross
-
compatib
ility with

device browser
s

and
as a means by which to
possibly
port to native iOS or Android devices
using
a tool

such as Appsify or
RubyMotion
.


We have ruled out developing
an iOS native application from
the
ground up because only a small
percentage of u
s have
an
adequate
amount of prior iOS development
experience and
only
two of us own
a
Mac (which is a prerequisite for
developing iOS applications).
Additionally,

the learning curve of
Objective
-
C is too steep to allow us
enough time to fully finish our a
pplication
on

time. Also, our team members do not
have expertise in

developing for the

Android platform.


9


However
,

we all h
ave experience with Javascript, and it

is a versatile
language

that major mobile operating systems utilize
. As long as the mobile dev
ice's
browser supports Javascript, our application
will

function

well

on the device.
Another perk is
that a
user do
es

not
need to
download our application from any
Market
place
/
Store; they can

simply go to our URL.


What is more, we discovered many o
pen source projects that allow
Javascript
-
based
application
s to

native
ly ported to the

iOS or Android platform
s

using well
-
established frameworks in a relatively short period of time.
T
his works
very well for a non
-
process
and in
-
app computation
intensive
application like ours.


Javascript applications are fast to develop thanks to the easy syntax, large
community, and well
-
supported documentation from Mozilla and Google. There are
also
many
more

readily available open source J
avascript libraries to choose
from
that
will help

us achieve certain functionalities.


We plan to include
only keystone features such as

i
nterview Q&A sharing
,

and basic social feature
s

as to

achieve satisfiable result
s

within
a
limited time
frame
.
We

plan to only implement very plain a
nd basic
social
-
centric
features that might
resemble or overlap with those from Facebook or Twitter, to avoid reinventing
the
wheel
. This allows us put more focus on our unique features to help people easily
share or browse interview questions to better ta
ckle their future interviews.

Last
but not least, we will employ third party login from Facebook or Google to simply
our user authentication features, which helps user easily connect with other who
are also using our application.